In the book, 40 experts speak, who explain in clear language what AI is, and what questions, challenges and opportunities the technology brings.
DOCUMENT
Explainable Artificial Intelligence (XAI) aims to provide insights into the inner workings and the outputs of AI systems. Recently, there’s been growing recognition that explainability is inherently human-centric, tied to how people perceive explanations. Despite this, there is no consensus in the research community on whether user evaluation is crucial in XAI, and if so, what exactly needs to be evaluated and how. This systematic literature review addresses this gap by providing a detailed overview of the current state of affairs in human-centered XAI evaluation. We reviewed 73 papers across various domains where XAI was evaluated with users. These studies assessed what makes an explanation “good” from a user’s perspective, i.e., what makes an explanation meaningful to a user of an AI system. We identified 30 components of meaningful explanations that were evaluated in the reviewed papers and categorized them into a taxonomy of human-centered XAI evaluation, based on: (a) the contextualized quality of the explanation, (b) the contribution of the explanation to human-AI interaction, and (c) the contribution of the explanation to human- AI performance. Our analysis also revealed a lack of standardization in the methodologies applied in XAI user studies, with only 19 of the 73 papers applying an evaluation framework used by at least one other study in the sample. These inconsistencies hinder cross-study comparisons and broader insights. Our findings contribute to understanding what makes explanations meaningful to users and how to measure this, guiding the XAI community toward a more unified approach in human-centered explainability.
MULTIFILE
Standard treatment for large burns is transplantation with meshed split skin autografts (SSGs). A disadvantage of this treatment is that healing is accompanied by scar formation. Application of autologous epidermal cells (keratinocytes and melanocytes) may be a suitable therapeutic alternative, since this may enhance wound closure and improve scar quality. A prospective, multicenter randomized clinical trial was performed in 40 adult patients with acute full thickness burns. On two comparable wound areas, conventional treatment with SSGs was compared to an experimental treatment consisting of SSGs in combination with cultured autologous epidermal cells (ECs) seeded in a collagen carrier. The primary outcome measure was wound closure after 5-7 days. Secondary outcomes were safety aspects and scar quality measured by graft take, scar score (POSAS), skin colorimeter (DermaSpectrometer) and elasticity (Cutometer). Wound epithelialization after 5-7 days was significantly better for the experimental treatment (71%) compared to the standard treatment (67%) (p = 0.034, Wilcoxon), whereas the take rates of the grafts were similar. No related adverse events were recorded. Scar quality was evaluated at 3 (n = 33) and 12 (n = 28) months. The POSAS of the observer after 3 and 12 months and of the patient after 12 months were significantly better for the experimental area. Improvements between 12% and 23% (p ≤ 0.010, Wilcoxon) were detected for redness, pigmentation, thickness, relief, and pliability. Melanin index at 3 and 12 months and erythema index at 12 months were closer to normal skin for the experimental treatment than for conventional treatment (p ≤ 0.025 paired samples t-test). Skin elasticity showed significantly higher elasticity (p = 0.030) in the experimental area at 3 months follow-up. We showed a safe application and significant improvements of wound healing and scar quality in burn patients after treatment with ECs versus SSGs only. The relevance of cultured autologous cells in treatment of extensive burns is supported by our current findings.
DOCUMENT
Both because of the shortcomings of existing risk assessment methodologies, as well as newly available tools to predict hazard and risk with machine learning approaches, there has been an emerging emphasis on probabilistic risk assessment. Increasingly sophisticated AI models can be applied to a plethora of exposure and hazard data to obtain not only predictions for particular endpoints but also to estimate the uncertainty of the risk assessment outcome. This provides the basis for a shift from deterministic to more probabilistic approaches but comes at the cost of an increased complexity of the process as it requires more resources and human expertise. There are still challenges to overcome before a probabilistic paradigm is fully embraced by regulators. Based on an earlier white paper (Maertens et al., 2022), a workshop discussed the prospects, challenges and path forward for implementing such AI-based probabilistic hazard assessment. Moving forward, we will see the transition from categorized into probabilistic and dose-dependent hazard outcomes, the application of internal thresholds of toxicological concern for data-poor substances, the acknowledgement of user-friendly open-source software, a rise in the expertise of toxicologists required to understand and interpret artificial intelligence models, and the honest communication of uncertainty in risk assessment to the public.
DOCUMENT
Recent advancements in mobile sensing and wearable technologies create new opportunities to improve our understanding of how people experience their environment. This understanding can inform urban design decisions. Currently, an important urban design issue is the adaptation of infrastructure to increasing cycle and e-bike use. Using data collected from 12 cyclists on a cycle highway between two municipalities in The Netherlands, we coupled location and wearable emotion data at a high spatiotemporal resolution to model and examine relationships between cyclists' emotional arousal (operationalized as skin conductance responses) and visual stimuli from the environment (operationalized as extent of visible land cover type). We specifically took a within-participants multilevel modeling approach to determine relationships between different types of viewable land cover area and emotional arousal, while controlling for speed, direction, distance to roads, and directional change. Surprisingly, our model suggests ride segments with views of larger natural, recreational, agricultural, and forested areas were more emotionally arousing for participants. Conversely, segments with views of larger developed areas were less arousing. The presented methodological framework, spatial-emotional analyses, and findings from multilevel modeling provide new opportunities for spatial, data-driven approaches to portable sensing and urban planning research. Furthermore, our findings have implications for design of infrastructure to optimize cycling experiences.
MULTIFILE
Previous research has shown clinical effectiveness of dermal substitution; however, in burn wounds, only limited effect has been shown. A problem in burn wounds is the reduced take of the autograft, when the substitute and graft are applied in one procedure. Recently, application of topical negative pressure (TNP) was shown to improve graft take. The aim of this study was to investigate if application of a dermal substitute in combination with TNP improves scar quality after burns. In a four-armed multicenter randomized controlled trial, a split-skin graft with or without a dermal substitute and with or without TNP was compared in patients with deep dermal or full-thickness burns requiring skin transplantation. Graft take and rate of wound epithelialization were evaluated. Three and 12 months postoperatively, scar parameters were measured. The results of 86 patients showed that graft take and epithelialization did not reveal significant differences. Significantly fewer wounds in the TNP group showed postoperative contamination, compared to other groups. Highest elasticity was measured in scars treated with the substitute and TNP, which was significantly better compared to scars treated with the substitute alone. Concluding, this randomized controlled trial shows the effectiveness of dermal substitution combined with TNP in burns, based on extensive wound and scar measurements.
DOCUMENT
High consumption of carbohydrates is linked to metabolic syndrome, possibly via the endogenous formation of advanced glycated end-products. Many Dutch elementary school children have a carbohydrate intake of >130g/day, the estimated minimum requirement. In this observational study, 126 Dutch elementary school children (5-12y of age) from two schools differing in frequency of gym lessons (2 or 5 times a week) were included. In all participants, height, weight, waist circumference, autofluorescence of skin glycated end-products (AGE-score), sports activity and carbohydrate consumption were recorded once. Sports activities in leisure time differentiated participants in ‘sportsmen’ and ‘non-sportsmen’. Carbohydrate intake and AGE score were positively associated in non-sportsmen (p<0.003), but negatively in sportsmen (p<0.002). In sportsmen, but not in non-sportsmen (p>0.50), a positive association was found (p<0.002) between carbohydrate intake and subject age. The intake of total carbohydrate and carbohydrates from juices and soft drinks was lower (p<0.001) at the Wassenberg School relative to the Alexander School. Based on waist to height ratio, >95% of the children had normal fat mass. No correlations were found between waist to height ratio or BMI and carbohydrate intake. Waist to height ratio was positively associated with BMI (p<0.001)) and subject age (p<0.001). Of all principal parameters, AGE score is most affected by being sportsmen or not (p<0.001). This study indicates that an increased intake of carbohydrates can be counteracted by sufficient physical activity (>2.5 hours per week). This implies that skin autofluorescence is a fast and non-invasive method to screen children for life style.
DOCUMENT
Spectral imaging has many applications, from methane detection using satellites to disease detection on crops. However, spectral cameras remain a costly solution ranging from 10 thousand to 100 thousand euros for the hardware alone. Here, we present a low-cost multispectral camera (LC-MSC) with 64 LEDs in eight different colors and a monochrome camera with a hardware cost of 340 euros. Our prototype reproduces spectra accurately when compared to a reference spectrometer to within the spectral width of the LEDs used and the ±1σ variation over the surface of ceramic reference tiles. The mean absolute difference in reflectance is an overestimate of 0.03 for the LC-MSC as compared to a spectrometer, due to the spectral shape of the tiles. In environmental light levels of 0.5 W m−2 (bright artificial indoor lighting) our approach shows an increase in noise, but still faithfully reproduces discrete reflectance spectra over 400 nm–1000 nm. Our approach is limited in its application by LED bandwidth and availability of specific LED wavelengths. However, unlike with conventional spectral cameras, the pixel pitch of the camera itself is not limited, providing higher image resolution than typical high-end multi- and hyperspectral cameras. For sample conditions where LED illumination bands provide suitable spectral information, our LC-MSC is an interesting low-cost alternative approach to spectral imaging.
MULTIFILE
The search for existing non-animal alternative methods for use in experiments is currently challenging because of the lack of both comprehensive structured databases and balanced keyword-based search strategies to mine unstructured textual databases. In this paper we describe 3Ranker, which is a fast, keyword-independent algorithm for finding non-animal alternative methods for use in biomedical research. The 3Ranker algorithm was created by using a machine learning approach, consisting of a Random Forest model built on a dataset of 35 million abstracts and constructed with weak supervision, followed by iterative model improvement with expert curated data. We found a satisfactory trade-off between sensitivity and specificity, with Area Under the Curve (AUC) values ranging from 0.85-0.95. Trials showed that the AI-based classifier was able to identify articles that describe potential alternatives to animal use, among the thousands of articles returned by generic PubMed queries on dermatitis and Parkinson's disease. Application of the classification models on time series data showed the earlier implementation and acceptance of Three Rs principles in the area of cosmetics and skin research, as compared to the area of neurodegenerative disease research. The 3Ranker algorithm is freely available at www.open3r.org; the future goal is to expand this framework to cover multiple research domains and to enable its broad use by researchers, policymakers, funders and ethical review boards, in order to promote the replacement of animal use in research wherever possible.
DOCUMENT
Background: Digital health is well-positioned in low and middle-income countries (LMICs) to revolutionize health care due, in part, to increasing mobile phone access and internet connectivity. This paper evaluates the underlying factors that can potentially facilitate or hinder the progress of digital health in Pakistan. Objective: The objective of this study is to identify the current digital health projects and studies being carried out in Pakistan, as well as the key stakeholders involved in these initiatives. We aim to follow a mixed-methods strategy and to evaluate these projects and studies through a strengths, weaknesses, opportunities, and threats (SWOT) analysis to identify the internal and external factors that can potentially facilitate or hinder the progress of digital health in Pakistan. Methods: This study aims to evaluate digital health projects carried out in the last 5 years in Pakistan with mixed methods. The qualitative and quantitative data obtained from field surveys were categorized according to the World Health Organization’s (WHO) recommended building blocks for health systems research, and the data were analyzed using a SWOT analysis strategy. Results: Of the digital health projects carried out in the last 5 years in Pakistan, 51 are studied. Of these projects, 46% (23/51) used technology for conducting research, 30% (15/51) used technology for implementation, and 12% (6/51) used technology for app development. The health domains targeted were general health (23/51, 46%), immunization (13/51, 26%), and diagnostics (5/51, 10%). Smartphones and devices were used in 55% (28/51) of the interventions, and 59% (30/51) of projects included plans for scaling up. Artificial intelligence (AI) or machine learning (ML) was used in 31% (16/51) of projects, and 74% (38/51) of interventions were being evaluated. The barriers faced by developers during the implementation phase included the populations’ inability to use the technology or mobile phones in 21% (11/51) of projects, costs in 16% (8/51) of projects, and privacy concerns in 12% (6/51) of projects.
DOCUMENT
