A comparative study on thyroid disease diagnosis using neural networks

Thyroid hormones produced by the thyroid gland help regulation of the body’s metabolism. Abnormalities of thyroid function are usually related to production of too little thyroid hormone (hypothyroidism) or production of too much thyroid hormone (hyperthyroidism). Thyroid disease diagnosis via proper interpretation of the thyroid data is an important classification problem. In this study, a comparative thyroid disease diagnosis were realized by using multilayer, probabilistic, and learning vector quantization neural networks. For this purpose, thyroid disease dataset which is taken from UCI machine learning database was used.     

Determination of Ni and Zn in an Oil Matrix using Schiff Base‐Assisted Extraction Followed by a Flame Atomic Absorption Spectrometer: A Simple Strategy to Determine Ni and Zn

In the present work, a new and simple Schiff base‐assisted extraction strategy for Ni and Zn from an edible oil matrix with subsequent determination using a flame atomic absorption spectrometer was suggested. According to the green approach, laborious sample‐pretreatment procedures were eliminated via complexation of the analytes with N,N′‐bis(4‐methoxysalycylidene)‐2‐hydroxy‐1,3‐propanediamine (4MSHP) and transferred from the oil phase to the aqueous phase. The complexation properties of 4MSHP, Ni, and Zn were investigated using UV–vis spectrophotometry. The experimental conditions that affect the extraction efficiency were optimized using central composite design. The optimum conditions for the extraction of Ni and Zn were as follows: a volume to oil mass ratio of 0.83 to 1.31 mL g^?1 of 4MSHP solution; 62.3‐ and 50.6‐min, stirring time; 27.3 and 31.1 °C, temperature, respectively. The detection limits (3s_bm^?1) were 0.41 μg g^?1 for Ni and 0.16 μg g^?1 for Zn. Validation of the suggested work was performed by the analysis of organometallic standard‐doped n‐hexane solutions as certified reference materials under the optimum experimental conditions. The recovery percentages were warranted the accuracy and found as 98.2 ± 1.8% for Ni and 99.8 ± 1.2% for Zn. In addition, relative SD values were below 5% for both the analytes. The Student's t‐test showed that there was no significant difference between the found and doped amount of analytes at 95% confidence level. The features such as the detection technique, cheapness, eco‐friendly solvent usage, and practicality were better compared to the literature.     

Improved machine learning performances with transfer learning to predicting need for hospitalization in arboviral infections against the small dataset

Neural Computing and Applications - The prediction of hospital patients and outpatients with suspected arboviral infection individuals in research-limited settings of the urban areas is defined as...     

Development of an electromagnetic hydrocyclone separator for purification of wastewater

A number of various organic and inorganic contaminants are commonly present in the technological fluids hindering production quality. Importantly, waste influents must be purified to meet the appropriate environmental standards for their disposal. This paper deals with the development of a treatment technique for technological and wastewater streams allowing the separation of microscopic droplets of water, sand, oil and other organic and inorganic contaminants. Some existing approaches and processes for treatment of contaminated fluids are briefly reviewed and the design of an electromagnetic hydrocyclone separator is presented in this paper. The effect of the main forces on the separation of small‐sized impurities (<10 μm) is analysed. The experimental tests demonstrated the efficiency of the developed apparatus in purifying the industrial wastewaters of the textile dyeing industry.     

Chaperone Therapy in Fabry Disease

Fabry disease is an X-linked lysosomal multisystem storage disorder induced by a mutation in the alpha-galactosidase A (GLA) gene. Reduced activity or deficiency of alpha-galactosidase A (AGAL) leads to escalating storage of intracellular globotriaosylceramide (GL-3) in numerous organs, including the kidneys, heart and nerve system. The established treatment for 20 years is intravenous enzyme replacement therapy. Lately, oral chaperone therapy was introduced and is a therapeutic alternative in patients with amenable mutations. Early starting of therapy is essential for long-term improvement. This review describes chaperone therapy in Fabry disease.     

Element fractionation analysis for infant formula and food additives by inductively coupled plasma optical emission spectrometry

Fractionation analyses are essential to investigate the effects of dietary constituents on bioavailability of elements. A sequential separation procedure has been applied for elemental fractionation patterns in infant formula, coffee cream, milk powder, whey powder and rice flour. The protein, lipid and serum fractions were separated successfully, and fractions were analysed for various essential (Fe, Cu, Zn, Mn, Mg, Ca, Cr and Co) and nonessential (Ni, V, Al, Mo, Pb, Sn, Ba and Cd) elements by ICP‐OES. Basically, three main fractions namely protein, lipid and serum were separated. Organically bounded fraction was calculated from the sum of the element contents in protein and lipid fractions. The organically bounded fraction can be retained longer in the body than the other fractions, and its percentages of whole elements are between 3.8% and 92.2% in the samples. Additionally, the distribution tendency of each studied metal was variable, which is based on the sample characteristics and complexation reactivity of the metal. The organically bounded fraction for Fe, Cu, Zn and Mn is higher than the other elements in whole samples except whey powder. Investigated elements are basically included in ionic forms in whey powder. Additionally, Mg and Ca are usually observed as uncomplexed structures in the samples.     

Spectrometric determination of iron and copper in vegetable oils after separation with Schiff base impregnated silica gel column: A simple approach for eliminating the high organic matrix

A simple solid‐phase extraction method has been described for the separation of Fe(III) and Cu(II) from liquid vegetable oils using N,N′‐bis(4‐methoxy salicylidene) ethylenediamine impregnated silica gel and the determination of these ions. The experimental parameters that affect the separation/preconcentration of ions were investigated by batch and column methods prior to the determination by flame atomic absorption spectrometry. Fe(III) and Cu(II) ions in 20.0 g portion of oil samples can be quantitatively sorbed and then eluted completely with 5.0 mL of 2.0 mol L^?1 HNO₃. Limits of detection were calculated as 22.8 and 13.9 μg kg^?1 for Fe(III) and Cu(II), respectively. The repetition of the suggested method was checked by finding relative standard deviation for five repeated analyses, which was 1.5% for Fe(III) and 0.1% for Cu(II). Applicability of the method was controlled with spiked and unspiked sunflower, corn, canola, olive, soya and hazelnut oils.     

Prenatal development and histochemical characteristics of gastrointestinal mucins in sheep fetuses

The object of this study was to describe the prenatal development and histochemical properties of mucins in the sheep gastrointestinal tract. To determine changes in the mucin profile, the sections were stained with specific histochemical stains for carbohydrates. While neutral and mixed mucins were observed in the superficial epithelial cells of the abomasal pyloric region, acidic mucins were detected in the secretory ducts and corpus of the glands. Acidic mucins consisted predominantly of sialomucins. In the duodenal villi, the number of goblet cells containing neutral mucins increased toward the end of gestation, whereas Brunner's glands contained acidic mucins until the 95th day of gestation and both acidic and neutral mucins thereafter. The jejunal goblet cells contained either acidic, neutral, or mixed mucins. Goblet cells containing acidic mucins, which were mainly localized to the ileal crypts and villi, mostly contained sulfated mucins. While villi were observed in the proximal colon until the 115th day of gestation, later the typical crypt structure emerged. During the period in which the villi were found in the proximal colon, the goblet cells containing sulphomucins were predominant, whereas the goblet cells containing sialomucins were predominant after the typical crypt structure was formed. In conclusion, gastrointestinal mucins may be involved in the formation of meconium during the prenatal period, and acidic mucins may contribute to the strength of the intestinal barrier against pathogens and digestive enzymes, as the barrier is not fully functional after birth.     

Diagnosis of chest diseases using artificial immune system

Chest diseases are one of the greatest health problems for people living in the developing world. Millions of people are diagnosed every year with a chest disease in the world. Chronic obstructive pulmonary, pneumonia, asthma, tuberculosis, lung cancer diseases are most important chest diseases and these are very common illnesses in Turkey. In this paper, a study on chest diseases diagnosis was realized by using artificial immune system. We obtained the classification accuracy with artificial immune system 93.84%. The result of the study was compared with the results of the previous similar studies reported focusing on chest diseases diagnosis. The chest diseases dataset were prepared from a chest diseases hospital’s database using patient’s epicrisis reports.     

An application of neural networks for harmonic coefficients and relative phase shifts detection

The varying the phase shifts will completely change the shape of the distorted wave, and may thus greatly affect the ability of the neural network to recognize harmonics. In this study, feed forward neural networks were used for the detection of the harmonic coefficients and relative phase shifts. The distorted wave including uniform distributed 5th, 7th, 11th, 13th, 17th, 19th, 23rd, 25th harmonics with up to 20° relative phase shifts were simulated and used. Two neural networks were used for this purpose. One of the neural networks was used for the detection of the 5th, 7th, 11th, 13th harmonic coefficients and the other was used for the detection of the relative phase shifts of these harmonics. Scaled conjugate gradient algorithm was used as training algorithm for the weights update of the neural networks. The results show that these neural networks are applicable to detect each harmonic coefficient and relative phase shift effectively.