Feature selection is used in machine learning to improve predictions, decrease computation time, reduce noise, and tune models based on limited sample data. In this article, we present FeatureExplorer, a visual analytics system that supports the dynamic evaluation of regression models and importance of feature subsets through the interactive selection of features in high-dimensional feature spaces typical of hyperspectral images. The interactive system allows users to iteratively refine and diagnose the model by selecting features based on their domain knowledge, interchangeable (correlated) features, feature importance, and the resulting model performance.

Objective: A high-fibre diet is associated with a lower risk for diseases. However, few adults meet the dietary fibre recommendation. Therefore, the effects and acceptance of an algorithm-generated personalised dietary advice (PDA) compared with general advice (GA) on fibre intake were investigated. Design: A 6-week, single-blind randomised controlled trial with a 3-month follow-up. Setting: PDA was based on habitual intake and provided fibre-rich alternatives using a website; GA contained brochures. Dietary intake was assessed at baseline, week 1, week 6 and 3-month follow-up. Both groups evaluated their advice at week 6. All participants had access to PDA from week 7 until 3-month follow-up. Participants: Two groups of healthy adults: PDA (n 34) and GA (n 47). For 3-month follow-up analysis, participants were re-divided into visitors (n 52) and non-visitors (n 26) of the PDA. Results: At week 6, energy intake remained stable in both groups, but fibre intake per 1000 kcal increased non-significantly in both groups (PDA = Δ0·5 ± 2·8; GA = Δ0·8 ± 3·1, P = 0·128). Importantly, a significantly higher percentage of PDA participants adhered to the recommendation compared with week 1 (PDA = 21 % increase; GA = 4 % increase, P ≤ 0·001). PDA participants evaluated the advice significantly better compared with GA participants. At 3-month follow-up, fibre intake increased compared with baseline (visitors = Δ2·2 ± 2·6, P < 0·001; non-visitors = Δ1·5 ± 1·9, P = 0·001), but was insignificantly different between groups. Visitors had a decrease and non-visitors had an increase in energy intake (visitors =Δ − 132 ± 525; non-visitors = Δ109 ± 507, P = 0·055). Conclusions: The algorithm-generated PDA was well accepted and stimulated adherence to the recommendations more than GA, indicating to be a suitable and cost-efficient method for improving dietary fibre intake in healthy adults.

No abstract

Background Irritable bowel syndrome (IBS) is the most prevalent functional bowel disorder, but its pathophysiology is still unknown. Although a microbial signature associated with IBS severity has been suggested, its association with IBS severity still remains largely unknown. Aims This study aims to assess longitudinal dynamics of fecal microbiota and short-chain fatty acids (SCFAs) in different IBS severity groups and study the association with stool pattern, diet, depression, anxiety, and quality of life (QoL). Methods A longitudinal study was performed, including n = 91 IBS patients and n = 28 matched controls. All participants collected fecal samples for microbiota composition and SCFA analysis and completed validated questionnaires regarding IBS severity, stool pattern, depression, anxiety, and IBS-QoL at two timepoints with four weeks in-between. Diet was assessed at the first timepoint. Results Over time, 36% of IBS patients changed in severity group, and 53% changed in predominant stool pattern. The largest proportion of microbiota variation was explained by the individual (R 2 = 70.07%). Microbiota alpha diversity and composition, and SCFAs did not differ between IBS severity groups, nor between IBS and controls. Relative abundances of Bifidobacterium, Terrisporobacter, and Turicibacter consistently differed between IBS and controls, but not between IBS severity groups. Large dynamics over time were observed in the association of microbiota composition with questionnaire data where IBS symptom severity was associated at T1 but not at T2. Conclusions Fecal microbiota and SCFA signatures were not consistently associated with IBS severity over time, indicating the importance of repeated sampling in IBS research.

In this study, the changes in microbial communities of lamb meat packaged in the air (plastic tray, PT) and in a vacuum pouch (VAC) were assessed by polymerase chain reaction-denaturing gradient gel electrophoresis (PCR-DGGE) during the storage at 4°C. For the PT lamb, the total viable count (TVC) was 10 7 CFU/g on Day 5, and the dominated bacteria were Pseudomonas fragi , P. fluorescens , and Acinetobacter spp. For the VAC lamb, the TVC was 10 7 CFU/g on Day 9, and the dominated bacteria were lactic acid bacteria, including Carnobacterium divergens , C. maltaromaticum , and Lactococcus piscium . One strain of Pseudomonas spp. also appeared in VAC lamb. The relative abundance of Enterobacteriaceae in VAC lamb was higher than that PT lamb, indicating a more important role of Enterobacteriaceae in spoilage for VAC lamb than that of PT lamb. The microbial compositions changed faster in the lamb stored in a PT than that stored in a VAC, and microbial community compositions of the late storage period were largely different from those of the early storage period for both the conditions. The findings of this study may guide improve the lamb hygiene and prolong the shelf life of the lamb.

Remote sensing platforms have become an effective data acquisition tool for digital agriculture. Imaging sensors onboard unmanned aerial vehicles (UAVs) and tractors are providing unprecedented high-geometric-resolution data for several crop phenotyping activities (e.g., canopy cover estimation, plant localization, and flowering date identification). Among potential products, orthophotos play an important role in agricultural management. Traditional orthophoto generation strategies suffer from several artifacts (e.g., double mapping, excessive pixilation, and seamline distortions). The above problems are more pronounced when dealing with mid- to late-season imagery, which is often used for establishing flowering date (e.g., tassel and panicle detection for maize and sorghum crops, respectively). In response to these challenges, this paper introduces new strategies for generating orthophotos that are conducive to the straightforward detection of tassels and panicles. The orthophoto generation strategies are valid for both frame and push-broom imaging systems. The target function of these strategies is striking a balance between the improved visual appearance of tassels/panicles and their geolocation accuracy. The new strategies are based on generating a smooth digital surface model (DSM) that maintains the geolocation quality along the plant rows while reducing double mapping and pixilation artifacts. Moreover, seamline control strategies are applied to avoid having seamline distortions at locations where the tassels and panicles are expected. The quality of generated orthophotos is evaluated through visual inspection as well as quantitative assessment of the degree of similarity between the generated orthophotos and original images. Several experimental results from both UAV and ground platforms show that the proposed strategies do improve the visual quality of derived orthophotos while maintaining the geolocation accuracy at tassel/panicle locations.

Hydrogenotrophic microbes play an essential role in the disposal of hydrogen and the maintenance of the hydrogen balance in gut ecosystems. Their abundances vary between individuals and have been reported to be associated with human gut disorders such as irritable bowel disease.

No abstract