Retina disease prediction using modified convolutional neural network based on Inception-ResNet model with support vector machine classifier

Artificial intelligence and deep learning have aided ocular disease through experiments including automatic illness recognition from images of the iris, fundus, or retina. Automated diagnosis systems (ADSs) provide services for the benefit of humanity and are essential in the early detection of harmful diseases. In fact, early detection is essential to avoid total blindness. In real life, several diagnostic tests such as visual ocular tonometry, retinal exam, and acuity test are performed, but they are conclusively time demanding and stressful for the patient. To consume time and detect the retinal disease earlier, an efficient prediction method is designed. In this proposed model, the first process is data collection that consists of a retinal disease dataset for testing and training. The second process is pre-processing, which executes image resizing and noise filter for feature extraction. The third step is feature extraction, which extracts the image's form, size, color, and texture for classification with CNN based on Inception-ResNet V2. The classification process is done by using the SVM with the extracted features. The prediction of diseases is classified such as normal, cataract, glaucoma, and retinal disease. The suggested model's performance is assessed using performance indicators such as accuracy, error, sensitivity, precision, and so forth. The suggested model's accuracy, error, sensitivity, and precision are 0.96, 0.962, 0.964, and 0.04, respectively, higher than existing techniques such as VGG16, Mobilenet V1, ResNet, and AlexNet. Thus, the proposed model instantly predicts retinal disease.     

A stochastic approximation approach to active queue management

Recently, a dynamic adaptive queue management with random dropping (AQMRD) scheme has been developed to capture the time-dependent variation of average queue size by incorporating the rate of change of average queue size as a parameter. A major issue with AQMRD is the choice of parameters. In this paper, a novel online stochastic approximation based optimization scheme is proposed to dynamically tune the parameters of AQMRD and which is also applicable for other active queue management (AQM) algorithms. Our optimization scheme significantly improves the throughput, average queue size, and loss-rate in relation to other AQM schemes.     

Activity of quassinoids as antifeedants against aphids

A series of quassinoids was tested for antifeedant activity against the aphidMyzus persicae (Hemiptera, Aphididae). Isobrucein B, brucein B and C, glaucarubinone, and quassin decreased feeding at concentrations down to 0.05% and isobrucein A was effective at 0.01%. Only quassin showed no phytotoxic effects and is therefore the most promising compound for further development.     

Accidental gamma dose measurement using commercial glasses

Commercial glasses have been investigated for their application in accidental gamma dose measurement using Thermoluminescent (TL) techniques. Some of the glasses have been found to be sensitive enough that they can be used as TL dating material in radiological accident situation for gamma dosimetry with lower detection limit 1 Gy (the dose significant for the onset of deterministic biological effects). The glasses behave linearly in the dose range 1-25 Gy with measurement uncertainty +/- 10%. The errors in accidental dose measurements using TL technique are estimated to be within +/- 25%. These glasses have shown TL fading in the range of 10-20% in 24 h after irradiation under room conditions; thereafter the fading becomes slower and reaches upto 50% in 15 d. TL fading of gamma-irradiated glasses follows exponential decay pattern, therefore dosimetry even after years is possible. These types of glasses can also be used as lethal dose indicator (3-4 Gy) using TL techniques, which can give valuable inputs to the medical professional for better management of radiation victims. The glasses are easy to use and do not require lengthy sample preparation before reading as in case of other building materials. TL measurement on glasses may give immediate estimation of the doses, which can help in medical triage of the radiation-exposed public.     

An ESCA study of the effectiveness of antiwear and extreme‐pressure additives based on substituted phosphorodithioate derivatives,and a comparison with ZDDP

Ashless substituted dithiophosphoric acid derivatives (ADPs) are a new generation of multifunctional additives with promising antiwear (AW) and extreme‐pressure (EP) characteristics. Three such additives synthesised in the authors' laboratory have been evaluated for their AW and EP properties by standard four‐ball friction and wear tests. The friction‐reducing properties of these additives were compared with those of a commercial zinc dialkyldithiophosphate (ZDDP). It was found that the phosphorodithioate compounds studied here possessed excellent AW/EP properties. Their AW characteristics were found to be comparable to those of ZDDP at low loads. However, at higher loads they show inferior AW characteristics in comparison to ZDDP. Nevertheless, ADP derived from cashew nut shell oil had a higher load‐carrying capacity than ZDDP. The mechanism of the AW and EP behaviour exhibited by the different additives was investigated using X‐ray photoelectron spectroscopy (XPS), Auger electron spectroscopy (AES), and scanning electron microscopy (SEM) of the worn surfaces formed during friction. XPS and AES analyses of the worn surfaces reveal that the tribochemical film formed on the ADP‐tested surfaces consisted mainly of metal phosphates and only a small amount of metal sulphides, even though the ADPs contained twice the number of sulphur atoms than phosphorus atoms. The ZDDP‐tested surface showed a mixture of metal sulphides and metal phosphates. Alkylamino substitution appeared to have no significant effect on the AW/EP properties of the additive. XPS and AES analyses also revealed that the tribochemical film formed on an ADP‐tested surface was thicker than that present on the ZDDP‐tested surface at low loads, whereas at higher loads the reverse was true. The higher weld load obtained for the blend containing cashew nut shell oil‐derived ADP is attributed to the thicker adsorbed reaction film formed on the surface due to the long alkyl groups present in the original additive structure. Short‐chain alkyl groups, however, form only a thin adsorbed layer, which may get rubbed off during the friction at high load. The low sulphide formation on ADP‐tested surfaces was attributed to the absence of any metal atom in the additive, which would help in the formation of metal sulphides during tribofragmentation and further tribochemical reactions.     

Determination of Calcium,Zinc, and Phosphorus in multigrade crankcase oils by XRF,DRES, and ICP-AES

The effect of viscosity index improvers such as styrene-isoprene copolymer (SICP), styrene-butadiene copolymer (SBCP), ethylene-propylene copolymer (OCP) and poly(alkyl) methacrylate (PMA), in multigrade oils on the determination of Ca, Zn and P in fresh and used lubricating oils was studied using XRF, DRES and ICP-AES techniques. Crankcase lubricating oil blends containing known amounts of Ca (2900 ppm), Zn (1700 ppm) and P (1600 ppm) were further formulated with different concentrations of these VI improvers and analysed for their Ca, Zn, and P contents by XRF, DRES and ICP-AES. Based on a comparison of the resulting analytical data, it was concluded that the VI improvers had no effect on the analysis results, and that the concentrations of Ca, Zn and P in the oil determined by these techniques correlate well with each other.     

Modification of Microstructure of Starch Extruded With Selected Lipids

Normal corn starch was extruded with palm, peanut and coconut oils; myristic, stearic and behenic acids and three monoglycerides in a C. W. Brabender single screw laboratory extruder at 140°C barrel temperature, 140rpm screw speed and 22% moisture content. Microstructure of extrudates was studied using a scanning electron mircocope. The pore size distribution, pore volumes and shear strengths of the extrudates were also studied. Addition of lipids resulted in modification of pore size distribution and pore volumes depending on the type of lipid. Complexing lipids gave lower pore volumes, smaller size pores, lower porosities and higher shear strengths of the extrudates. Porosity correlated well with the shear strength values.     

Effects of nano size mischmetal and its oxide on improving the hydrogen sorption behaviour of MgH2

This paper reports the catalytic effects of mischmetal (Mm) and mischmetal oxide (Mm-oxide) on improving the dehydrogenation and rehydrogenation behaviour of magnesium hydride (MgH₂). It has been found that 5 wt.% is the optimum catalyst (Mm/Mm-oxide) concentration for MgH₂. The Mm and Mm-oxide catalyzed MgH₂ exhibits hydrogen desorption at significantly lower temperature and also fast rehydrogenation kinetics compared to ball-milled MgH₂ under identical conditions of temperature and pressure. The onset desorption temperature for MgH₂ catalyzed with Mm and Mm-oxide are 323 °C and 305 °C, respectively. Whereas the onset desorption temperature for the ball-milled MgH₂ is 381 °C. Thus, there is a lowering of onset desorption temperature by 58 °C for Mm and by 76 °C for Mm-oxide. The dehydrogenation activation energy of Mm-oxide catalyzed MgH₂ is 66 kJ/mol. It is 35 kJ/mol lower than ball-milled MgH₂. Additionally, the Mm-oxide catalyzed dehydrogenated Mg exhibits faster rehydrogenation kinetics. It has been noticed that in the first 10 min, the Mm-oxide catalyzed Mg (dehydrogenated MgH₂) has absorbed up to 4.75 wt.% H₂ at 315 °C under 15 atmosphere hydrogen pressure. The activation energy determined for the rehydrogenation of Mm-oxide catalyzed Mg is ∼62 kJ/mol, whereas that for the ball-milled Mg alone is ∼91 kJ/mol. Thus, there is a decrease in absorption activation energy by ∼29 kJ/mol for the Mm-oxide catalyzed Mg. In addition, Mm-oxide is the native mixture of CeO₂ and La₂O₃ which makes the duo a better catalyst than CeO₂, which is known to be an effective catalyst for MgH₂. This takes place due to the synergistic effect of CeO₂ and La₂O₃. It can thus be said that Mm-oxide is an effective catalyst for improving the hydrogen sorption behaviour of MgH₂.