Explainable AI: Machine Learning Interpretation in Blackcurrant Powders.

Recently, explainability in machine and deep learning has become an important area in the field of research as well as interest, both due to the increasing use of artificial intelligence (AI) methods and understanding of the decisions made by models. The explainability of artificial intelligence (XAI) is due to the increasing consciousness in, among other things, data mining, error elimination, and learning performance by various AI algorithms. Moreover, XAI will allow the decisions made by models in problems to be more transparent as well as effective. In this study, models from the 'glass box' group of Decision Tree, among others, and the 'black box' group of Random Forest, among others, were proposed to understand the identification of selected types of currant powders. The learning process of these models was carried out to determine accuracy indicators such as accuracy, precision, recall, and F1-score. It was visualized using Local Interpretable Model Agnostic Explanations (LIMEs) to predict the effectiveness of identifying specific types of blackcurrant powders based on texture descriptors such as entropy, contrast, correlation, dissimilarity, and homogeneity. Bagging (Bagging_100), Decision Tree (DT0), and Random Forest (RF7_gini) proved to be the most effective models in the framework of currant powder interpretability. The measures of classifier performance in terms of accuracy, precision, recall, and F1-score for Bagging_100, respectively, reached values of approximately 0.979. In comparison, DT0 reached values of 0.968, 0.972, 0.968, and 0.969, and RF7_gini reached values of 0.963, 0.964, 0.963, and 0.963. These models achieved classifier performance measures of greater than 96%. In the future, XAI using agnostic models can be an additional important tool to help analyze data, including food products, even online.

Efficiency of Identification of Blackcurrant Powders Using Classifier Ensembles.

In the modern times of technological development, it is important to select adequate methods to support various food and industrial problems, including innovative techniques with the help of artificial intelligence (AI). Effective analysis and the speed of algorithm implementation are key points in assessing the quality of food products. Non-invasive solutions are being sought to achieve high accuracy in the classification and evaluation of various food products. This paper presents various machine learning algorithm architectures to evaluate the efficiency of identifying blackcurrant powders (i.e., blackcurrant concentrate with a density of 67 °Brix and a color coefficient of 2.352 (E520/E420) in combination with the selected carrier) based on information encoded in microscopic images acquired via scanning electron microscopy (SEM). Recognition of blackcurrant powders was performed using texture feature extraction from images aided by the gray-level co-occurrence matrix (GLCM). It was evaluated for quality using individual single classifiers and a metaclassifier based on metrics such as accuracy, precision, recall, and F1-score. The research showed that the metaclassifier, as well as a single random forest (RF) classifier most effectively identified blackcurrant powders based on image texture features. This indicates that ensembles of classifiers in machine learning is an alternative approach to demonstrate better performance than the existing traditional solutions with single neural models. In the future, such solutions could be an important tool to support the assessment of the quality of food products in real time. Moreover, ensembles of classifiers can be used for faster analysis to determine the selection of an adequate machine learning algorithm for a given problem.

The Need for Machines for the Nondestructive Quality Assessment of Potatoes with the Use of Artificial Intelligence Methods and Imaging Techniques.

This article describes chemical and physical parameters, including their role in the storage, trade, and processing of potatoes, as well as their nutritional properties and health benefits resulting from their consumption. An analysis of the share of losses occurring during the production process is presented. The methods and applications used in recent years to estimate the physical and chemical parameters of potatoes during their storage and processing, which determine the quality of potatoes, are presented. The potential of the technologies used to classify the quality of potatoes, mechanical and ultrasonic, and image processing and analysis using vision systems, as well as their use in applications with artificial intelligence, are discussed.

Predictive Models of Phytosterol Degradation in Rapeseeds Stored in Bulk Based on Artificial Neural Networks and Response Surface Regression.

The need to maintain the highest possible levels of bioactive components contained in raw materials requires the elaboration of tools supporting their processing operations, starting from the first stages of the food production chain. In this study, artificial neural networks (ANNs) and response surface regression (RSR) were used to develop models of phytosterol degradation in bulks of rapeseed stored under various temperatures and water activity conditions (T = 12-30 °C and aw = 0.75-0.90). Among ANNs, networks based on a multilayer perceptron (MLP) and a radial basis function (RBF) were tested. The model input constituted aw, temperature and storage time, whilst the model output was the phytosterol level in seeds. The ANN-based modeling turned out to be more effective in estimating phytosterol levels than the RSR, while MLP-ANNs proved to be more satisfactory than RBF-ANNs. The approximation quality of the ANNs models depended on the number of neurons and the type of activation functions in the hidden layer. The best model was provided by the MLP-ANN containing nine neurons in the hidden layer equipped with the logistic activation function. The model performance evaluation showed its high prediction accuracy and generalization capability (R2 = 0.978; RMSE = 0.140). Its accuracy was also confirmed by the elliptical joint confidence region (EJCR) test. The results show the high usefulness of ANNs in predictive modeling of phytosterol degradation in rapeseeds. The elaborated MLP-ANN model may be used as a support tool in modern postharvest management systems.

The Process of Pasting and Gelling Modified Potato Starch with LF-NMR.

Currently, society expects convenience food, which is healthy, safe, and easy to prepare and eat in all conditions. On account of the increasing popularity of modified potato starch in food industry and its increasing scope of use, this study focused on improving the physical modification of native starch with temperature changes. As a result, it was found that the suggested method of starch modification with the use of microwave power of 150 W/h had an impact on the change in starch granules. The LF-NMR method determined the whole range of temperatures in which the creation of a starch polymer network occurs. Therefore, the applied LF-NMR technique is a highly promising, noninvasive physical method, which allows obtaining a better-quality structure of potato starch gels.

Deep and Machine Learning Using SEM, FTIR, and Texture Analysis to Detect Polysaccharide in Raspberry Powders.

In the paper, an attempt was made to use methods of artificial neural networks (ANN) and Fourier transform infrared spectroscopy (FTIR) to identify raspberry powders that are different from each other in terms of the amount and the type of polysaccharide. Spectra in the absorbance function (FTIR) were prepared as well as training sets, taking into account the structure of microparticles acquired from microscopic images with Scanning Electron Microscopy (SEM). In addition to the above, Multi-Layer Perceptron Networks (MLPNs) with a set of texture descriptors (machine learning) and Convolution Neural Network (CNN) with bitmap (deep learning) were devised, which is an innovative attitude to solving this issue. The aim of the paper was to create MLPN and CNN neural models, which are characterized by a high efficiency of classification. It translates into recognizing microparticles (obtaining their homogeneity) of raspberry powders on the basis of the texture of the image pixel.

Application of Deep and Machine Learning Using Image Analysis to Detect Fungal Contamination of Rapeseed.

This paper endeavors to evaluate rapeseed samples obtained in the process of storage experiments with different humidity (12% and 16% seed moisture content) and temperature conditions (25 and 30 °C). The samples were characterized by different levels of contamination with filamentous fungi. In order to acquire graphic data, the analysis of the morphological structure of rapeseeds was carried out with the use of microscopy. The acquired database was prepared in order to build up training, validation, and test sets. The process of generating a neural model was based on Convolutional Neural Networks (CNN), Multi-Layer Perceptron Networks (MLPN), and Radial Basis Function Networks (RBFN). The classifiers that were compared were devised on the basis of the environments Tensorflow (deep learning) and Statistica (machine learning). As a result, it was possible to achieve the lowest classification error of 14% for the test set, 18% classification error for MLPN, and 21% classification error for RBFN, in the process of recognizing mold in rapeseed with the use of CNN.

Simultaneous Bilateral Minimally Invasive Direct Anterior Approach Total Hip Arthroplasty with fast track Protocol.

BACKGROUND: Advanced degenerative hip joint disease is bilateral in approximately 20% of cases, prompting questions of whether it is necessary to perform two separate surgical procedures, whether simultaneous bilateral hip replacement makes the surgical treatment too extensive, and whether it significantly affects the postoperative course. MATERIAL AND METHODS: The study analysed the duration of hospitalisation, perioperative complications, and the need for blood transfusion in 30 patients (27 men and 3 women) with bilateral hip osteoarthritis who underwent simultaneous bilateral total hip replacement from a minimally invasive direct anterior approach followed by a fast track protocol for optimisation of perioperative management between 2014 and 2017. The mean age of patients was 60.2 years (range 43 to 77 years) and the mean follow-up period was 28 months (range 18 to 48 months). RESULTS: Mean duration of hospitalisation was 4.5 days (range 3 to 9 days). A total of 4 patients (13%) required allogeneic blood transfusion. No patient developed thromboembolic or infectious complications or implant dislocation after surgery. Apart from one case where the acetabulum was not selected correctly, which resulted in postoperative loosening, there were no other significant medical events potentially related to the surgical treatment. CONCLUSIONS: Simultaneous bilateral total hip arthroplasty using a minimally invasive direct approach and a fast track protocol for optimisation of perioperative management does not increase the need for perioperative blood transfusion or the number of surgical complications and constitutes a safe, effective, and recommendable method of treatment in patients with advanced bilateral degenerative disease of the hip joints.

Classification of Dried Strawberry by the Analysis of the Acoustic Sound with Artificial Neural Networks.

In this paper, the authors used an acoustic wave acting as a disturbance (acoustic vibration), which travelled in all directions on the whole surface of a dried strawberry fruit in its specified area. The area of space in which the acoustic wave occurs is defined as the acoustic field. When the vibrating surface-for example, the surface of the belt-becomes the source, then one can observe the travelling of surface waves. For any shape of the surface of the dried strawberry fruit, the signal of travelling waves takes the form that is imposed by this irregular surface. The aim of this work was to research the effectiveness of recognizing the two trials in the process of convection drying on the basis of the acoustic signal backed up by neural networks. The input variables determined descriptors such as frequency (Hz) and the level of luminosity (dB). During the research, the degree of crispiness relative to the degree of maturity was compared. The results showed that the optimal neural model in respect of the lowest value of the root mean square turned out to be the Multi-Layer Perceptron network with the technique of dropping single fruits into water (data included in the learning data set Z2). The results confirm that the choice of method can have an influence on the effectives of recognizing dried strawberry fruits, and also this can be a basis for creating an effective and fast analysis tool which is capable of analyzing the degree of ripeness of fruits including their crispness in the industrial process of drying fruits.

The Use of Image Analysis to Detect Seed Contamination-A Case Study of Triticale.

Samples of triticale seeds of various qualities were assessed in the study. The seeds were obtained during experiments, reflecting the actual sowing conditions. The experiments were conducted on an original test facility designed by the authors of this study. The speed of the air (15, 20, 25 m/s) transporting seeds in the pneumatic conduit was adjusted to sowing. The resulting graphic database enabled the distinction of six classes of seeds according to their quality and sowing speed. The database was prepared to build training, validation and test sets. The neural model generation process was based on multi-layer perceptron networks (MLPN) and statistical (machine training). When the MLPN was used to identify contaminants in seeds sown at a speed of 15 m/s, the lowest RMS error of 0.052 was noted, whereas the classification correctness coefficient amounted to 0.99.

The Use of Lignin as a Microbial Carrier in the Co-Digestion of Cheese and Wafer Waste.

The aim of the article was to present the effects of lignin grafted with polyvinylpyrrolidone (PVP) as a microbial carrier in anaerobic co-digestion (AcoD) of cheese (CE) and wafer waste (WF). Individual samples of waste cheese and wafers were also tested. The PVP modifier was used to improve the adhesive properties of the carrier surface. Lignin is a natural biopolymer which exhibits all the properties of a good carrier, including nontoxicity, biocompatibility, porosity, and thermal stability. Moreover, the analysis of the zeta potential of lignin and lignin combined with PVP showed their high electrokinetic stability within a wide pH range, that is, 4-11. The AcoD process was conducted under mesophilic conditions in a laboratory by means of anaerobic batch reactors. Monitoring with two standard parameters: pH and the VFA/TA ratio (volatile fatty acids-to-total alkalinity ratio) proved that the process was stable in all the samples tested. The high share of N-NH4+ in TKN (total Kjeldahl nitrogen), which exceeded 90% for WF+CE and CE at the last phases of the process, proved the effective conversion of nitrogen forms. The microbiological analyses showed that eubacteria proliferated intensively and the dehydrogenase activity increased in the samples containing the carrier, especially in the system with two co-substrates (WF+CE/lignin) and in the waste cheese sample (CE/lignin). The biogas production increased from 1102.00 m3 Mg-1 VS (volatile solids) to 1257.38 m3 Mg-1 VS in the WF+CE/lignin sample, and from 881.26 m3 Mg-1 VS to 989.65 m3 Mg-1 VS in the CE/lignin sample. The research results showed that the cell immobilization on lignin had very positive effect on the anaerobic digestion process.

Neural Image Analysis and Electron Microscopy to Detect and Describe Selected Quality Factors of Fruit and Vegetable Spray-Dried Powders-Case Study: Chokeberry Powder.

The study concentrates on researching possibilities of using computer image analysis and neural modeling in order to assess selected quality discriminants of spray-dried chokeberry powder. The aim of the paper is the quality identification of chokeberry powders on account of their highest dying power, the highest bioactivity, as well as technologically satisfying looseness of the powder. The article presents neural models with vision techniques backed up by devices such as digital cameras, as well as an electron microscope. The reduction in size of input variables with PCA has an influence on improving the processes of learning data sets, thus increasing the effectiveness of identifying chokeberry fruit powders included in digital pictures, which is shown in the results of the conducted research. The effectiveness of image recognition is presented by classifying abilities, as well as low Root Mean Square Error (RMSE), for which the best results are achieved with a typology of network type Multi-Layer Perceptron (MLP). The selected networks type MLP are characterized by the highest degree of classification at 0.99 and RMSE at 0.11 at most at the same time.

Structure and corrosion resistance of Co-Cr-Mo alloy used in Birmingham Hip Resurfacing system.

PURPOSE: The endoprostheses made of cobalt-chromium-molybdenum (Co-Cr-Mo) alloys belong to the group of the most popular metallic implants used for reconstruction of hip joints. For such biomaterials, the primary goal is a correct and long-term functioning in the aggressive environment of body fluids. Therefore, the purpose of this study was to examine both the morphology and the corrosion resistance of implants made of the cobalt alloy used in Birmingham Hip Resurfacing (BHR) system (Smith & Nephew). For comparative purposes, the electrochemical studies were done for the nitrided stainless steel - Orthinox. METHODS: Observations of the microstructure of the material under investigation were performed by means of the optical metallographic microscope and the scanning electron microscope. Furthermore, Energy Dispersive X-ray Spectroscopy was used to analyse the chemical composition of the endoprosthesis. Characterisation and evaluation of electrochemical corrosion resistance of the selected alloys were performed by potentiodynamic polarisation tests. RESULTS: The structural studies confirmed that Co-Cr-Mo (BHR system) is characterised by a typical dendritic microstructure with carbide precipitates, mainly M23C6, within the interdendritic areas. The results of the polarisation measurements showed that the cobalt alloy investigated exhibits lower corrosion potential than Orthinox in the utilised environments (3% NaCl, simulated body fluid - Hank's Body Fluid). CONCLUSIONS: However, the high passivation ability of the Co-Cr-Mo alloy, as well as its resistance to the initiation and propagation of localised corrosion processes, indicate that this material is significantly more appropriate for long-term implants.