This study aimed to investigate the effect of four green extraction techniques (ultrasound-assisted extraction, UAE; supercritical fluid extraction, SFE; subcritical water extraction, SWE; and extraction using deep eutectic solvents, DES) on the extraction of targeted flavonoids from edible feijoa flowers. The bioactive components in the obtained extracts were quantified by High-Performance Liquid Chromatography—Photodiode Array Detector (HPLC-PDA). Moreover, total polyphenol content and antioxidant activity by DPPH•, ABTS•+, FRAP, and CUPRAC assays were investigated. UAE generally gave the highest yields for isoquercitrin and quercetin content (18.36–25.33 and 10.86–16.13 µg/g), while DES extraction with choline chloride:lactic acid (1:2) and H2O content of 50% gave the highest yield of chrysanthemin (90.81 µg/g). The highest yield of flavone (12.69 mg/g) was obtained with supercritical CO2 at 300 bar. Finally, UAE gave the highest total polyphenol content (ca. 64 mg GAE/g) and antioxidant activity at 70 °C during 30 min with 40% (0.84 mmol TEAC/g and 2.25 mmol Fe2+/g, for ABTS•+ and CUPRAC, respectively) and 60% ethanol-water solution (0.49 mmol TEAC/g and 2.09 mmol Fe2+/g, for DPPH• and FRAP, respectively). The eco-friendly extraction techniques resulted in selective methods capable of extracting targeted bioactive compounds from edible feijoa flowers.

Intelligent manufacturing is becoming increasingly important due to the growing demand for maximizing productivity and flexibility while minimizing waste and lead times. This work investigates automated secondary robotic food packaging solutions that transfer food products from the conveyor belt into containers. A major problem in these solutions is varying product supply which can cause drastic productivity drops. Conventional rule-based approaches, used to address this issue, are often inadequate, leading to violation of the industry's requirements. Reinforcement learning, on the other hand, has the potential of solving this problem by learning responsive and predictive policy, based on experience. However, it is challenging to utilize it in highly complex control schemes. In this paper, we propose a reinforcement learning framework, designed to optimize the conveyor belt speed while minimizing interference with the rest of the control system. When tested on real-world data, the framework exceeds the performance requirements (99.8% packed products) and maintains quality (100% filled boxes). Compared to the existing solution, our proposed framework improves productivity, has smoother control, and reduces computation time.

In modern cardiology, sodium‐glucose cotransporter 2 (SGLT2) inhibitors are critical components of heart failure (HF) treatment algorithms and exert their effects primarily by preventing glucose reabsorption and facilitating its urinary excretion. The objective was to systematically review randomized controlled trials (RCTs) assessing the effects of SGLT2 inhibitors, particularly canagliflozin, empagliflozin, dapagliflozin, ertugliflozin, sotagliflozin (dual SGLT inhibitor), and their use in HF. Systematic searches of PubMed/Medline, The Cochrane Central Register of Controlled Trials (CENTRAL), and ClinicalTrials.gov databases were performed. There were no restrictions imposed on the date and status of publication; however, there were restrictions on language for the searched studies. A total of 1139 records were identified in the bibliographic searches from both databases and the register of choice for this systematic review. Following duplicate removal, screening for titles and abstracts, and thorough assessment of full‐text articles, 12 RCTs met the inclusion criteria. Altogether, 83 878 patients were included in this review. Among the included studies, two RCTs, with six respective reports, investigated canagliflozin, four RCTs with 13 derived reports investigated dapagliflozin, three RCTs with 12 separate reports studied the effects of empagliflozin, one RCT and its three respective reports assessed ertugliflozin's effects, and two RCTs with one added report investigated the dual inhibitor sotagliflozin. Pooled meta‐analytic effects of SGLT2 inhibitors were as follows: on atrial fibrillation odds ratio (OR) = 0.83, 95% confidence interval (CI): 0.68–1.01, prediction interval (PI): 0.57–1.19; on HF hospitalization OR = 0.69, 95% CI: 0.60–0.78, PI: 0.60–0.78; on cardiovascular death OR = 0.82, 95% CI: 0.58–1.15, PI: 0.42–1.60; and on major adverse cardiovascular events OR = 0.90, 95% CI: 0.77–1.06, PI: 0.71–1.15. SGLT2 inhibitors significantly improve the quality of life in HF patients. Their beneficial effects on HF, especially in left ventricular dysfunction, have made their use possible irrespective of diabetes mellitus or atrial fibrillation status.

This paper investigated the bending moment of chair base joints. The ultimate bending moments (maximum moment), calculated on the base of the measured maximum applied loads (maximum force), were compared for the front leg and rear leg joints of a chair base. The joints had different angles between the stretcher and the leg (joint angle) as well different tenon lengths (30 mm and 32 mm). The results of the tests indicated that for different test specimen configurations but the same tenon-and-mortise geometry, the maximum force of joints with a smaller value of joint angle (front leg joints) was higher than the force values of joints with a larger angle (rear leg joints) for all tenon lengths. However, the results showed less difference among the calculated bending moments of the analysed sets of joints. A significant difference was not revealed between the bending moments of joints with a smaller value of joint angle and the bending moments of joints with a larger angle for all tenon lengths. A significant difference between the bending moments for the tenon length of 30 mm and tenon length of 32 mm was determined for rear leg joints but not for front leg joints. The presented approach of joint strength analysis through the testing of specimens with different shapes and dimensions are applicable to research and practice.

In this paper, the elastic behaviour in bending of three-layer plywood cantilever beams is analysed. Deflections of straight and half-circle cantilevers, loaded with a force at free end is determined experimentally and calculated using analytical and finite element method approach. The analytical calculation of deflection for the strait cantilever is obtained using a transformed cross section. The deflection of half-circle cantilever is determined by the classical laminated plate theory and Castigliano’s theorem. Loads and cantilever dimensions are varied in the study using the design of experiment. The deflection regression models for straight and semicircular plywood cantilevers are obtained from the experimental results. Analytically and numerically determined deflections of strait and half-circle cantilevers show very good agreement. Experimentally recorded deflections are approximately 30 % higher than analytical values. Stiffness properties and deflection values are influenced by direction of fibres in the outer layers of a three-layer plywood beam.

Recent vision architectures and self-supervised training methods enable vision models that are extremely accurate and general, but come with massive parameter and computational costs. In practical settings, such as camera traps, users have limited resources, and may fine-tune a pretrained model on (often limited) data from a small set of specific categories of interest. These users may wish to make use of modern, highly-accurate models, but are often computationally constrained. To address this, we ask: can we quickly compress large generalist models into accurate and efficient specialists? For this, we propose a simple and versatile technique called Few-Shot Task-Aware Compression (TACO). Given a large vision model that is pretrained to be accurate on a broad task, such as classification over ImageNet-22K, TACO produces a smaller model that is accurate on specialized tasks, such as classification across vehicle types or animal species. Crucially, TACO works in few-shot fashion, i.e. only a few task-specific samples are used, and the procedure has low computational overheads. We validate TACO on highly-accurate ResNet, ViT/DeiT, and ConvNeXt models, originally trained on ImageNet, LAION, or iNaturalist, which we specialize and compress to a diverse set of"downstream"subtasks. TACO can reduce the number of non-zero parameters in existing models by up to 20x relative to the original models, leading to inference speedups of up to 3$\times$, while remaining accuracy-competitive with the uncompressed models on the specialized tasks.

: Satureja montana L. (SM) has a long traditional use as a spice and a medicine for various gastrointestinal disorders, including painful spasms and diarrhea. Contrary to conventional drugs, administration of SM and its extracts are considered safe. Previous studies have shown that the essential oils (EOs) of SM from different areas are rich in monoterpenes, sesquiterpenes, diterpens, and phenolic compounds, including flavonoids, tannins, and acids with great composition variability. Determination of composition of EO from Herzegovinian SM done by gas chromatography-flame ionization detection and gas chromatography mass spectrometry (GC-FID and GC/MS, respectively) revealed carvacrol as a primary substance followed by γ -terpinene, p -cymene, and β -caryophyllene. Ex vivo spasmolitic activity caused by EO was evident in different types of isolated rat ileum function with the most potent effect on spontaneous activity followed by electrical field stimulation and KCl-and CaCl 2 -induced activity. SMEO produced in vivo antidiarreal activity on castor oil-induced diarrhea in young rats and showed the potential to cause a decrese water content in the feces of adult Wistar rats.This study indicates that effects of SM on the intestinum could be mediated through combination of Kv channel activation and Ca channel blockade, but additional mechanisms might be involved. The results of this study corroborate the traditional use of SM as antispasmodic, antidiarrheal, and antisecretory agents.

Background: The clinical course of patients with incomplete reperfusion after thrombectomy, defined as an expanded Thrombolysis in Cerebral Infarction (eTICI) score of 2a–2c, is heterogeneous. Patients showing delayed reperfusion (DR) have good clinical outcomes, almost comparable to patients with ad-hoc TICI3 reperfusion. We aimed to develop and internally validate a model that predicts DR occurrence in order to inform physicians about the likelihood of a benign natural disease progression. Patients and methods: Single-center registry analysis including all consecutive, study-eligible patients admitted between 02/2015 and 12/2021. Preliminary variable selection for the prediction of DR was performed using bootstrapped stepwise backward logistic regression. Interval validation was performed with bootstrapping and the final model was developed using a random forests classification algorithm. Model performance metrics are reported with discrimination, calibration, and clinical decision curves. Primary outcome was concordance statistics as a measure of goodness of fit for the occurrence of DR. Results: A total of 477 patients (48.8% female, mean age 74 years) were included, of whom 279 (58.5%) showed DR on 24 follow-up. The model’s discriminative ability for predicting DR was adequate (C-statistics 0.79 [95% CI: 0.72–0.85]). Variables with strongest association with DR were: atrial fibrillation (aOR 2.06 [95% CI: 1.23–3.49]), Intervention-To-Follow-Up time (aOR 1.06 [95% CI: 1.03–1.10]), eTICI score (aOR 3.49 [95% CI: 2.64–4.73]), and collateral status (aOR 1.33 [95% CI: 1.06–1.68]). At a risk threshold of R = 30%, use of the prediction model could potentially reduce the number of additional attempts in one out of four patients who will have spontaneous DR, without missing any patients who do not show spontaneous DR on follow-up. Conclusions: The model presented here shows fair predictive accuracy for estimating chances of DR after incomplete thrombectomy. This may inform treating physicians on the chances of a favorable natural disease progression if no further reperfusion attempts are made.

Banja Luka is a city which, in October 1969, experienced the strongest earthquake in the territory of Bosnia and Herzegovina. On that occasion 2/3 of the school buildings were severely damaged or destroyed. Banja Luka students were forced to attend that school year throughout the former Yugoslavia. Seismic risk management for schools on the territory of Banja Luka is significant from the aspect of protecting students as a vulnerable category of society. It is also important observing the fact that school facilities are used as facilities for temporary mass accommodation of the affected population during emergencies. Examining the level of earthquake protection, i.e. seismic resilience of the schools in Banja Luka, as a city with a high seismic hazard, implies the resistance of school buildings (material resilience) but also the preparedness of school communities (administration, students and teachers) to react properly in the event of an earthquake (non-material resilience). The results of the research indicate weaknesses in both the material and non-material resilience of schools. The structural aspects of school resilience include the seismic hazard of the area, the soil at school locations in terms of the expected seismic effect, the age and poor maintenance of the buildings, and the undefined ownership of school buildings. Regarding non-structural resistance, a low level of carrying out preventive activities such as education, training and practical exercises for dealing with earthquakes was identified as well as inadequate planning documentation.

Unreinforced unconfined solid brick masonry walls were experimentally tested in full scale (233x241x25cm) and reduced scale (100x100x25cm) at the laboratory of the Institute for materials and structures, Faculty of Civil Engineering in Sarajevo. Cantilever walls were loaded in cyclic shear or pushed monotonically. In order to study the nonlinear behavior in a detailed and global manner, finite element meso- and macro-models of the tested walls were created using the finite element software Diana FEA. Brick units are discretized by continuum elements in a meso-model and discontinuity in displacement field is introduced by interface elements between units. In order to account for brick cracking, an additional interface element was added in the unit middle. Continuum macro-models approximate heterogeneous masonry wall by a single material and discretization is independent of brick layout, i.e., bricks, mortar and unit-mortar interface are smeared out in the continuum. The recently developed engineering masonry model is an orthotropic total-strain continuum model with smeared cracking and it was used with shell elements. Numerical results are verified against the data obtained from the experimental research program. The walls exhibit rocking failure mode in low precompression, while diagonal cracking occurs for higher vertical stresses. The results show good matching with the experimentally obtained curves regarding the ultimate load and ductility.

This paper presents the methodology for seismic analysis of masonry structures that can be employed in commercial software packages such as SAP2000. The concept of elementary block which combines non-linear spring and linear shell elements is used for discretization of masonry walls. The proposed modelling technique with localized nonlinearity can successfully simulate in-plane wall failure modes induced by compressive or tensile axial force and transverse force. It can also be used to investigate out-of-plane collapse which makes it a good candidate for 3D static and dynamic analysis of buildings. The modelling approach is tested on two examples where pushover analysis was performed: a single slender cantilever masonry wall and a family house. The response was verified against the results delivered by 3MURI and MINEA, and reasonable agreement was obtained. It is demonstrated that the transverse walls have significant contribution to the load bearing capacity of buildings.