Adaptation and application of multi-objective evolutionary algorithms for rule reduction and parameter tuning of fuzzy rule-based systems

Recently, multi-objective evolutionary algorithms have been applied to improve the difficult tradeoff between interpretability and accuracy of fuzzy rule-based systems. It is known that both requirements are usually contradictory, however, these kinds of algorithms can obtain a set of solutions with different trade-offs. This contribution analyzes different application alternatives in order to attain the desired accuracy/interpr-etability balance by maintaining the improved accuracy that a tuning of membership functions could give but trying to obtain more compact models. In this way, we propose the use of multi-objective evolutionary algorithms as a tool to get almost one improved solution with respect to a classic single objective approach (a solution that could dominate the one obtained by such algorithm in terms of the system error and number of rules). To do that, this work presents and analyzes the application of six different multi-objective evolutionary algorithms to obtain simpler and still accurate linguistic fuzzy models by performing rule selection and a tuning of the membership functions. The results on two different scenarios show that the use of expert knowledge in the algorithm design process significantly improves the search ability of these algorithms and that they are able to improve both objectives together, obtaining more accurate and at the same time simpler models with respect to the single objective based approach.

---

The Green language type system

A programming language that considers basic values and classes as objects brings more opportunities of code reuse and it is easier to use than a language that does not support this feature. However, popular statically typed object-oriented languages do not consider classes as first-class objects because this concept is difficult to integrate with static type checking. They also do not consider basic values as objects for sake of efficiency. This article presents the Green language type system which supports classes as classless objects and offers a mechanism to treat basic values as objects. The result is a reasonably simple type system which is statically typed and easy to implement. It simplifies several other language mechanisms and prevents any infinite regression of metaclasses.     

Immunoglobulin 1 (IgG1) Fc-glycosylation profiling of anti-citrullinated peptide antibodies from human serum

In several autoimmune disorders, including rheumatoid arthritis (RA), autoantibodies are thought to be the driving force of pathogenicity. Glycosylation of the Fc-part of human Igs is known to modulate biological activity. Hitherto, glycosylation of human IgG-Fc has been analyzed predominantly at the level of total serum IgG, revealing reduced galactosylation in RA. Given the pathogenic relevance of autoantibodies in RA, we wished, in the present study, to address the question whether distinct Fc-glycosylation features are observable at the level of antigen-specific IgG subpopulations. For this purpose, we have developed a method for the microscale purification and Fc-glycosylation analysis of anti-citrullinated peptide antibodies (ACPA). ACPA represent a group of autoantibodies that occur with unique specificity in RA patients. Their presence is associated with increased inflammatory disease activity and rapid joint destruction. Results indicate that ACPA of the IgG1 subclass vary considerably from total serum IgG1 with respect to Fc-galactosylation, with galactosylation being higher on ACPA than on serum IgG1 for some patients, while other patients show higher galactosylation on serum IgG1 than on ACPA. Using this method, studies can be performed on the biological and clinical relevance of ACPA glycosylation within RA patient cohorts.     

Proteomic analysis in cerebrospinal fluid of patients with atypical nonketotic hyperglycinemia and pulmonary hypertension - A pilot study

A variant phenotype of nonketotic hyperglycinemia has been described by our group associated with pulmonary hypertension. The aim of this study is to investigate the cerebrospinal fluid proteomes to get an insight into this neurodegenerative process producing leukoencephalopathy with white matter spongiform degeneration. DIGE and MALDI-TOF-TOF analyses were performed to carry out the proteomic study of four patients against three normal controls and one additional control of a classical nonketotic hyperglycinemia. The differential proteomic analysis showed a displacement of some series of spots toward the acidic side. The shifted proteins showed a high degree of carbonylation and increased methionine sulfoxidation was found in cystatin C and in vitamin-D-binding protein. These findings in addition to the increase of serum malondialdehyde concentration provide evidence of an oxidative stress in the patients under study, which is probably systemic rather than mainly confined to the CNS. The similarities of our findings with those found in other neurodegenerative diseases suggest that oxidative damage is commonly involved in these pathologies. DIGE technology improves the 2-D PAGE differential analysis and it is suitable in proteomic studies with a small number of cases.     

Post-embryonic development of the antennal sensilla in Melipona quadrifasciata anthidioides (Hymenoptera: Meliponini)

The sensilla are sensory organs formed by cuticular and cellular structures specialized in reception of chemical and physical stimuli from the environment and transmission to the insect's central nervous system. In function of the great concentration of sensilla, the antennae are the main organs for interaction between bees and with the environment. This work studied the presence of antennal sensilla in the different phases of pupal development of the stingless bee Melipona quadrifasciata anthidioides by means of scanning electron microscopy and light microscopy. The results showed that antennal sensilla begin their development in the transition of the prepupae to the white-eyed pupae and finish it in the pigmented-body pupae phase. The antennal sensilla were exposed to the environment in the black-eyed pupae when the old cuticle is completely digested, suggesting that only in the final pupal phases can these bees perceive the environmental stimuli.     

Intervention practices in musculoskeletal disorder prevention: a critical literature review

Musculoskeletal disorders (MSDs) affect much of the workforce and remain a major form of occupational ill health. With a view to improving the efficacy of interventions, this review examined preventative actions relating to these disorders. A detailed analysis grid was used to classify the information contained in the 47 reviewed articles whose common aspect was to report actions carried out in the workplace that led to the implementation of changes to prevent MSDs. The analysis identified and characterized three major categories of intervention processes in MSD prevention: the complete type (n=17), the shortened type (n=16), and the turnkey type (n=14). These three groups of intervention processes were differentiated by their approaches and their contexts of application. The result was important differences in the changes implemented. Because of the variability in intervention processes and possible impacts on MSD prevention, a proposal to "delimit" these processes so as to improve their effectiveness is presented.     

Effects of double-muscling on carcass quality, beef tenderness and myofibrillar protein degradation in Belgian Blue White bulls

Carcass properties and meat quality characteristics of 32 Belgian Blue White double-muscled bulls (DM) were compared with those of 59 bulls of the same breed with normal conformation (N). DM showed superior carcass quality as revealed in increased dressing percentage, meat production yield, conformation grade, muscle/fat and muscle/bone ratios (all P < 0·001). Longissimus dorsi (LD) shear force values, drip and cooking losses at 8 days post mortem (pm) were significantly higher (P < 0·001) for DM, whereas sarcomere lengths were not different. Calpain 1 and calpastatin levels at 1 h and 24 h pm were tremendously decreased in DM as were also cathepsin B and L levels at 1 and 8 days pm (P < 0·001). As evident from semi-quantitative SDS-PAGE, these differences were accompanied by higher titin and lower 30 kDa levels (P < 0·001) in DM. Troponin-t levels were not different, but very low. Intramuscular collagen content was significantly lower in DM (P < 0·001). This suggests that lower background toughness in DM was compensated for by reduced post-mortem proteolytic tenderization. Discrepancy with literature reports regarding tenderness of DM might be related to the extreme muscularity of the Belgian Blue White breed, compared to other DM breeds. The results also suggest that reduced protein turnover might be involved in the muscle hypertrophy phenomenon within this breed, because of likely reduced levels of calpains and cathepsins in living DM animals.     

Field cross correlator for analysis of ultrafast signals

The cross-correlation function between two light fields is recorded with the help of a new device. The proposed correlator exhibits ultrashort time resolution. The optical path difference between the two interfering beams does not have to be known with interferometric precision. The experimental dynamic range proved to be as large as 10(5). The device features imaging capabilities that could be applied to the analysis of two-dimensional images with ultrashort time resolution.     

Objective meat quality measurements of ham: A practical classification method on the slaughterline

Meat quality is one of the principal problems in Spanish dry cured ham productuon. At the experimental level, there are well developed objective instrumental measurements which can be made on the slaughterline and then, by applying the correct statistical treatment, these measurements can be used to classify hams with an acceptable degree of accuracy. These treatments, however, are not easy to apply in the industry.

In this work, 312 pork carcasses were tested at three industrial abbatoirs. Ham meat quality was determined by internal light scattering (FOP), electrical conductivity (EC) and pH measurements in Semimembranosus (SM) muscle at 45 min and 24 h post mortem. The results obtained were: 21 (6.7%) DFD; 168 (53.9%) slightly DFD; 104 (33.3%) normal; 11 (3.5%) slightly PSE; and 8 (2.6%) PSE.

A simplified method has been developed for classifying hams into quality categories, by deriving a theoretical quality value (Q) based on a simple statistical treatment of values for FOP, EC and pH. The multiple linear regression coefficient of Q was R² = 0.85 (P<0.001) with respect to the experimental variables.     

Improving robustness of industrial object detection by automatic generation of synthetic images from CAD models

Object detection (OD) is used for visual quality control in factories. Images that compose training datasets are often collected directly from the production line and labeled with bounding boxes manually. Such data represent well the inference context but might lack diversity, implying a risk of overfitting. To address this issue, we propose a dataset construction method based on an automated pipeline, which receives a CAD model of an object and returns a set of realistic synthetic labeled images (code publicly available). Our approach can be easily used by non-expert users and is relevant for industrial applications, where CAD models are widely available. We performed experiments to compare the use of datasets obtained by the two different ways—collecting and labeling real images or applying the proposed automated pipeline—in the classification of five different industrial parts. To ensure that both approaches can be used without deep learning expertise, all training parameters were kept fixed during these experiments. In our results, both methods were successful for some objects but failed for others. However, we have shown that the combined use of real and synthetic images led to better results. This finding has the potential to make industrial OD models more robust to poor data collection and labeling errors, without increasing the difficulty of the training process.