Identification of natural food-derived emulsifiers using QSAR and machine learning: Application in dairy emulsions.

We investigated the effect of κ-carrageenan (κ-CG) addition and supplementation with Na-CN, WPI, and MPI on the rheological properties, microstructure, fat globule size distribution, and stability of low-fat cream (LFC) emulsions containing 19% milk fat. Rheological parameter (ηa,10, K, and G′) values of LFCs increased with κ-CG addition. At 0.1% κ-CG concentration, MPI-stabilized LFC had higher G′ value (64.8 Pa) than those of WPI-stabilized LFC (51.3 Pa) and Na-CN-stabilized LFC (34.8 Pa), indicating the strong interaction between MPI and κ-CG. Cryo-SEM images showed the protein layers on milk fat globule membrane in dairy emulsions and revealed that κ-CG led to depletion-flocculation of milk fat globules and promoted clustering of globules. The creaming index values of LFCs with 0.10% κ-CG were lower than those of LFCs with 0.05% κ-CG. Thus, high κ-CG concentrations can contribute to the physical stability in dairy emulsions stabilized by milk proteins. Practical applications Milk proteins are widely used as emulsifiers, and κ-carrageenan (κ-CG) is added to dairy emulsion products to improve their physical properties and stability. Therefore, in this study, the effect of κ-CG addition on physical properties and stability of low-fat creams (LFCs) supplemented with different milk protein types (Na-CN, WPI, and MPI) was investigated. The strong synergistic interaction between proteins and κ-CG was observed in MPI-stabilized LFCs when compared to Na-CN- and WPI-stabilized LFCs. We found that the presence of κ-CG in MPI-stabilized LFCs improved the stability of LFCs by inhibiting the phase separation due to the high viscosity of the continuous phase. These findings suggest that dairy cream-based food products with desirable physical properties can be realized through κ-CG addition to LFC emulsions with milk proteins such as ice cream mix, coffee creamer, and light cream.

Olive leaves are obtained as a waste product of the olive industry. Biophenols, abundantly found in olive leaves, are susceptible to heat, light, and oxidizing agents, which necessitates encapsulation to increase their bioavailability. In this study, the double emulsion method was preferred due to its protective effect on the active substance and the control over its release. The effects of different pea flour concentrations (15%, 20%, and 25%) used in outer aqueous phase of double emulsion and homogenization methods [high-speed homogenization (HSH) and ultrasonication (US)] on emulsion properties were investigated. The particle size, rheology, encapsulation efficiency, stability, optical images, and release behavior of the emulsions were determined. As hypothesized, flours acted as emulsifiers in the outer aqueous phase to increase the stability of emulsions. It was observed that the stability of emulsions was correlated with the viscosity and particle size. Increasing pea flour concentration from 15% to 25% resulted in a 25% and 30% increase in the stability of double emulsions prepared with HSH and US, respectively. The higher stability of emulsions prepared with 25% was due to their higher viscosity and smaller particle size. Samples were found to have shear-thinning behavior. Moreover, emulsions stored at 20 °C showed faster degradation compared to 4 °C. US treatment did not decrease the average particle size of emulsions. Average encapsulation efficiency for double emulsions prepared with HSH and US was 88.3% and 85.9%, respectively. As a result, pea flours could be used to encapsulate olive leaf extract successfully with high encapsulation efficiencies by using the double emulsion method.

The effect of emulsifiers, emulsion stabilizer (maltodextrin, MD), and β-cyclodextrin (BCD) on physical and oxidative properties of oil-in-water (O/W) emulsions (5%, 20%, 40% of oil, w/w) was investigated. Four different emulsifiers were selected based on their structure: two types of protein-based emulsifiers (sodium caseinate, SC; and whey protein isolate, WPI), and two types low molecular weight emulsifiers (LMEWs: lecithin, LEC; and Citrem, CITREM). Physical and oxidative stability of emulsions prepared with these emulsifiers together with MD were compared based on their creaming index (CI), viscosity, droplet size, zeta potential, peroxide and p-anisidine values. LMWE-stabilized emulsions (with LEC or CITREM) had better creaming stability with lower droplet sizes whereas protein-stabilized emulsions (with SC or WPI) had higher viscosities. Droplet size was the lowest when CITREM was used, which increased with increasing oil concentration for all emulsifiers. Formulation with the lowest CI value and droplet size was considered to be more prone to oxidation; therefore, a 1:1 (w/w) combination of CITREM with BCD was used to stabilize the emulsions to improve the oxidative as well as physical stability. Added BCD significantly increased the storage stability of emulsions by reducing CI and droplet size values with a simultaneous increase in the viscosity, both at room temperature and at storage conditions (at 4 and 55 o C). However, the oxidative as well as physical stability of BCD added emulsions were not improved, neither toward heat- nor light-induced lipid oxidation. PRACTICAL APPLICATION: This work investigated the effects of emulsifiers and dextrins on the stability of oil-in-water (O/W) emulsions. Both maltodextrin (MD) and β-cyclodextrin (BCD) addition resulted in enhanced physical stability, the latter being more effective. The findings can be applied to formulate emulsions with improved shelf life within the limits of allowed daily intake (ADI) level of BCD (5 mg/kg bw per day).

Impact of pea protein-inulin conjugates prepared via the Maillard reaction using a combination of ultrasound and pH-shift treatments on physical and oxidative stability of algae oil emulsions

The objective of this work was to compare the physical stability and physicochemical properties of emulsions, containing enzymatically modified fatty base and homogenized mechanically or by ultrasounds. In the study, lipase‐catalyzed interesterification of mutton tallow and hemp seed oil, in a ratio of 3:1, 3:3, and 1:3 w/w, was performed in order to produce fatty bases of the emulsions. Reaction conditions were selected to obtain increased amount of the by‐products (MAG and DAG), which were applied as the only emulsifiers in dispersion systems. The higher ratio of animal fat in the interesterified fatty basis of an emulsion had an impact on the greater stability of these systems. The correlation between thickening agent concentration in the prepared emulsions and stability was not observed. Smaller particle size was found in emulsions manufactured by ultrasonic homogenization, although it did not contribute to greater long‐term stability of these emulsions. It was concluded that emulsions containing enzymatically modified fats and homogenized mechanically revealed greater physical stability than their counterparts homogenized with ultrasounds.

The present paper reports on the use of phenolic extracts from olive mill wastewater (OMW) in model olive oil-in-water (O/W) emulsions to study their effect on their physical and chemical stability. Spray-dried OMW polyphenols were added to a model 20% olive O/W emulsion stabilized with whey protein isolate (WPI) and xanthan gum, in phosphate buffer solution at pH 7. The emulsions were characterised under accelerated storage conditions (40 °C) up to 30 days. Physical stability was evaluated by analysing the creaming rate, mean particle size distribution and mean droplet size, viscosity and rheological properties, while chemical stability was assessed through the measurement of primary and secondary oxidation products. The rheological behaviour and creaming stability of the emulsions were dramatically improved by using xanthan gum, whereas the concentration of WPI and the addition of encapsulated OMW phenolics did not result in a significant improvement of physical stability. The formation of oxidation products was higher when higher concentrations of encapsulated polyphenols were used, indicating a possible binding with the WPI added in the system as a natural emulsifier. This paper might help in solving the issue of using the olive mill wastewater from olive processing in formulating functional food products with high antioxidant activity and improved health properties.

Despite the obvious benefits of double emulsions in reducing fat content by replacing it with the water phase, their physical and oxidative stability remains a major concern. The objective of this study was to determine the ability of black chokeberry extract to inhibit lipid oxidation during storage at 4 °C for 60 days when different amounts of the extract were added to the inner water phase of the double emulsion. In the first step, the physical stability of the emulsions was evaluated. Higher amount of the extract caused the formation of double emulsions with smaller droplets and higher viscosity. Throughout the whole storage period, the double emulsions showed good physical stability and high encapsulation efficiency (EE) of the extract (>95%) in the inner water phase. The positive effect of the extract on the oxidative stability of the double emulsions was shown by measuring changes in peroxide values and conjugated dienes and through the Oxipres and Rancimat tests during the convenient and accelerated storage of emulsions for 60 days. The higher amount of extract suppressed lipid oxidation to a higher extent given the significant amount of polyphenolics in the extract. © 2021 Society of Chemical Industry.

Plant-based polyphenols are increasingly being explored as functional ingredients in emulsified food systems. In this study, the effects of sesamol on the physical and chemical stability of flaxseed oil-in-water emulsions stabilized by either phospholipids (sunflower) or proteins (whey or pea) were investigated. In the absence of sesamol, the protein-based emulsions displayed better physical stability than the phospholipid-based ones, which was related to their smaller particle diameter and higher particle charge. For the phospholipid-based emulsions, sesamol addition did not improve their physical stability, but it did inhibit lipid oxidation. In particular, it decreased the formation of secondary oxidation products, with a 65% reduction in TBAR formation compared to the control after 8 days of storage. For the protein-based emulsions, sesamol addition reduced particle aggregation and inhibited lipid oxidation, reducing the secondary oxidation products by around 85% after 19 days of storage. The inhibitory efficiency of sesamol in the pea protein-based emulsions was comparable to that in the whey protein-based ones. The effects of sesamol on the physical and chemical stability of the emulsions were related to its partitioning between the oil, water, and interfacial layers. This study suggests that adding sesamol to plant-based emulsions may improve their physical and chemical stability, thereby extending their shelf life.

Comparing the effectiveness of natural and synthetic emulsifiers on oxidative and physical stability of avocado oil-based nanoemulsions

Stability, rheological properties and microstructure of Pickering emulsions stabilized by different concentration of glidian/sodium caseinate nanoparticles using konjac glucomannan as structural regulator

This study aims to assess the performance of machine learning (ML) techniques in optimising nurse staffing and evaluating the appropriateness of nursing care delivery models in hospital wards. The primary outcome measures include the adequacy of nurse staffing and the appropriateness of the nursing care delivery system. Historical and current healthcare challenges, such as nurse shortages and increasing patient acuity, necessitate innovative approaches to nursing care delivery. For instance, the COVID-19 pandemic highlighted the need for flexible and scalable staffing models to manage surges in patient volume and acuity. A descriptive study was conducted in 39 inpatient wards across a university hospital and three state hospitals, involving 117 ward-level observations. Data were collected using the Rush Medicus Patient Classification Scale and analysed using k-Nearest Neighbour, Support Vector Machine, Random Forest, and Logistic Regression algorithms. Effectiveness was measured by the accuracy of machine learning predictions regarding nurse staffing adequacy, while suitability was determined by the congruence between observed nursing care models and patient needs. STROBE checklist. The Random Forest algorithm demonstrated the highest accuracy in predicting both nurse staffing adequacy and the appropriateness of nursing care delivery systems. The study found that 68.4% of wards had sufficient nurse staffing and 26.5% of wards used appropriate care delivery models, with functional nursing and total patient care models being the most commonly used. The study highlights functional nursing and total patient care models, emphasising the need to consider nurse qualifications and patient needs in selecting care systems. Machine learning, particularly the Random Forest algorithm, proved effective in aligning staffing with patient requirements. Machine learning, particularly the Random Forest algorithm, proves effective in optimising nursing care delivery models, suggesting significant potential for enhancing patient care and nurse satisfaction. The research underscores machine learning's role in improving nursing care delivery, aligning nurse staffing with patient needs, and advancing healthcare outcomes. The findings advocate for integrating machine learning in the planning of nursing care delivery models. This study sets a precedent for using data-driven approaches to improve nurse staffing and care delivery, potentially enhancing global clinical outcomes and operational efficiencies. The global clinical community can learn from this study the value of employing machine learning techniques to make informed, evidence-based decisions in healthcare management. While the study lacked direct patient involvement, its goal was to enhance patient care and healthcare efficiency. Future research will aim to incorporate patient and public insights more directly.

Small particle size lipid emulsions, satiety and energy intake in lean men

The effects of traffic-derived air pollution can be controlled through the provision of adequate and effective air quality control and mitigation measures with the aid of air quality models. This paper examines the application of Machine Learning (ML) methods (Random Forests (RF), Extreme Learning Machine (ELM) and Deep Learning (DL)) algorithms in air quality modelling. Data collected from continuous monitoring stations in London comprising air pollutants, traffic and meteorological variables were used for training the models. The selected ML methods were trained to predict roadside PM10 and PM2.5 concentrations. The results obtained showed that all the methods can be used for training models for the prediction of the PM10 and PM2.5 concentrations. The RF, ELM and DL algorithms were found to be suitable for this purpose due to their predictive accuracy and faster training speed. The models performed slightly better in predicting PM2.5 than the PM10 concentrations. The average performance of the machine learning models was found to be 0.90 and 0.94 (R); 99% and 98% (FAC2); 9.2 and 4.5 (RMSE) and 0.83 and 0.84(IOA) for the prediction of PM10, PM2.5. The advantages of using the Deep Learning, ELM and RF algorithms over traditional ANN are the faster training speed and scalability especially when using high-performance computing.

Türkiye’de ikinci el araç piyasası her zaman hareketli olmuştur. İkinci el araç piyasasında marka, model, yakıt türü gibi özelliklerin ne kadar yoğunlukta olduğu, ne kadar fiyata etki ettiği gibi faktörler analiz edilerek, bu bilgiler kullanışlı hale getirilebilir. Araçların çeşitli özelliklerine göre fiyatları değişmektedir. Fiyatları tahmin edebilmek için makine öğrenme teknikleri kullanılabilir ve kullanıcıların araç satarken veya alırken fiyat belirlemelerine yardımcı olabilir. Fiyat tahmini, veri madenciliğinin bir görevi olan fonksiyon tahmini veya regresyon sınıfına girmektedir. İkinci el araç sayısı oldukça fazla olduğundan dolayı bu çalışmada analizler yapılırken büyük veri sistemleri kullanılmıştır. Apache Spark ve makine öğrenme kütüphanesi bunun için oldukça kullanışlıdır. Fiyat tahmini için doğrusal regresyon, karar ağacı regresyonu, rastgele orman regresyonu, GBT regresyonu, izotonik regresyon algoritmaları kullanılmıştır. Kullanılan algoritmalar ile araçların fiyat tahmini yapılmıştır ve en yüksek başarıyı 21435,09 RMSE ve 0,887 R2 değerleriyle rastgele orman algoritması elde etmiştir. Rasgele orman algoritması ve diğer algoritmalarla elde edilen RMSE ve R2 değerleri arasında anlamlı bir farklılık olup olmadığını kontrol için yapılan istatistiksel testler sonucunda, rasgele orman algoritması ile elde edilen sonuçların daha iyi olduğu sonucuna ulaşılmıştır. Rasgele orman algoritmasının daha iyi sonuçlar vermesinin nedeni, algoritmanın birden çok karar ağacı üzerinden eğitim gerçekleştirmesi, esnekliği ve güçlü hiper parametrelere sahip olmasıdır.

There is robust evidence that heart failure (HF) is associated with substantial mortality, morbidity, poor health-related quality of life, healthcare utilization, and economic burden. Previous research has revealed that there are sex differences in the epidemiology, etiology, and disease burden of HF. However, research on HF among women, especially postmenopausal women, is limited. To fill the knowledge gap, the three related aims of this dissertation were to: (1) identify knowledge gaps in HF research among women, especially postmenopausal women, using unsupervised machine learning methods and big data (i.e., articles published in PubMed); (2) identify emerging predictors (i.e., polypharmacy and some prescription medications) of incident HF among postmenopausal women using supervised machine learning methods; (3) identify leading predictors of HF-related emergency room use among postmenopausal women using supervised machine learning methods with data from a large commercial insurance claims database in the United States. This study utilized machine learning methods. In the first aim, non-negative matrix factorization algorithms were used to cluster HF articles based on the primary topic. Clusters were independently validated and labeled by three investigators familiar with HF research. The most understudied area among women was atrial fibrillation. Among postmenopausal women, the most understudied topic was stress-induced cardiomyopathy. For the second and third aims, a retrospective cohort design and Optum’s de-identified Clinformatics® Data Mart Database (Optum, Eden Prairie, MN), de-identified health insurance claims data, were used. In the second aim, multivariable logistic regression and three classification machine learning algorithms (cross-validated logistic regression (CVLR), random forest (RF), and eXtreme Gradient Boosting (XGBoost) algorithms) were used to identify predictors of incident HF among postmenopausal women. The associations of the leading predictors to incident HF were explored with an interpretable machine learning SHapley Additive exPlanations (SHAP) technique. The eight leading predictors of incident HF consistent across all models were: older age, arrhythmia, polypharmacy, Medicare, chronic obstructive pulmonary disease (COPD), coronary artery disease, hypertension, and chronic kidney disease. Some prescription medications such as sulfonylureas and antibiotics other than fluoroquinolones predicted incident HF in some machine learning algorithms. In the third aim, a random forest algorithm was used to identify predictors of HF-related emergency room use among postmenopausal women. Interpretable machine learning techniques were used to explain the association of leading predictors to HF-related emergency room use. Random forest algorithm had high predictive accuracy in the test dataset (Area Under the Curve: 94%, sensitivity: 93%, specificity: 77%, and accuracy: 0.81). We found that