Yazar "Abut, Serdar" seçeneğine göre listele

Listeleniyor 1 - 13 / 13
  • [ X ]
    Öğe
    AI-Based Model Design for Prediction of COPD Grade from Chest X-Ray Images: A Model Proposal (COPD-GradeNet)
    (2024) Abut, Serdar
    Chronic Obstructive Pulmonary Disease (COPD) ranks high among the leading causes of death, particularly in middle- and low-income countries. Early diagnosis of COPD is challenging, with limited diagnostic methods currently available. In this study, a artificial intelligence model named COPD-GradeNet is proposed to predict COPD grades from radiographic images. However, the model has not yet been tested on a dataset. Obtaining a dataset including spirometric test results and chest X-ray images for COPD is a challenging process. Once the proposed model is tested on an appropriate dataset, its ability to predict COPD grades can be evaluated and implemented. This study may guide future research and clinical applications, emphasizing the potential of artificial intelligence-based approaches in the diagnosis of COPD.
  • [ X ]
    Öğe
    Anaerobic co-digestion of oil-extracted spent coffee grounds with various wastes: Experimental and kinetic modeling studies
    (Elsevier Sci Ltd, 2021) Atelge, M. R.; Atabani, A. E.; Abut, Serdar; Kaya, M.; Eskicioglu, Cigdem; Semaan, Georgeio; Lee, Changsoo
    The effect of oil extraction from spent coffee grounds as a pre-treatment strategy prior to anaerobic digestion besides assessing the feasibility of defatted spent coffee grounds co-digestion with spent tea waste, glycerin, and macroalgae were examined. Mesophilic BMP tests were performed using defatted spent coffee grounds alongside four co-substrates in the ratio of 25, 50, and 75%, respectively. The highest methane yield was obtained with the mono-digestion of defatted spent coffee grounds with 336 +/- 7 mL CH4/g VS and the yield increased with the increase in the mass ratio of defatted spent coffee grounds during co-digestion. Moreover, defatted spent coffee grounds showed the highest VS and TS removal at 35.5% and 32.1%, respectively and decreased thereafter. Finally, a linear regression model for the interaction effects between substrates was demonstrated and showed that distinctly mixing defatted spent coffee grounds, spent coffee grounds, and spent tea waste outperforms other triple mixed substrates.
  • [ X ]
    Öğe
    Comparative investigation of multi-walled carbon nanotube modified diesel fuel and biogas in dual fuel mode on combustion, performance, and emission characteristics
    (Elsevier Sci Ltd, 2022) Atelge, M. R.; Arslan, Esenay; Krisa, David; Al-Samaraae, R. R.; Abut, Serdar; unalan, Sebahattin; Atabani, A. E.
    Biogas has been investigated as an alternative biofuel in dual fuel operating mode in a direct injection diesel engine. However, there is not sufficient information about using modified fuels with biogas. This study aimed to investigate the effects of modified diesel fuel and biogas on combustion behavior, performance, and emissions characteristics at 1500 rpm constant speed with 5 different load conditions at an interval of 25%. Diesel was modified with multi-walled carbon nanotubes with 30, 60, and 90 ppm. Diesel fuel and three modified fuels were used as pilot fuel and biogas was introduced through the intake manifold with the flow rate of 500 g/h as the primary fuel. Diesel mode fuels were denominated F1 while dual fuel mode fuels were labeled as F2, and the concentration levels were given subscript such as F2 (@60ppm). The experimental study revealed that modified fuel showed better combustion behaviors, performance, and emissions in comparison to diesel fuel. Further, the same trend was observed in the dual fuel mode. The maximum pressure of F2(@60 ppm) was 1% higher than F2 under dual fuel mode at the full load. Moreover, the coefficient of variation of the indicated mean effective pressure for dual fuel mode was found approximately 9.2, 6.9, 6.2, and 7.2% for F2, F2(@30 ppm), F2(@60 ppm), and F2(@90 ppm), respectively at full load. In addition, the energy share of biogas increased by 7.9, 8.7, and 7.1% for F2(@30 ppm), F2(@60 ppm), and F2(@90 ppm), respectively in comparison with F2 at full load. The highest decrease of brake specific energy consumption under the dual mode was obtained to be an 8% drop from F2(@60 ppm) compared to F2 at full load. At the same load, the brake thermal efficiency of F2(@30 ppm), F2(@60 ppm), and F2(@90 ppm) were noted to be 30.2, 30.4, and 30.0%, respectively which are higher than F2 (27.9%). The value of replaced diesel with biogas was noted 0.09, 0.23, 0.24, and 0.22 kg/h for F2, F2(@30 ppm), F2(@60 ppm), and F2(@90 ppm), respectively under the full load condition. Lastly, CO and HC emissions were almost the same value with and without modified fuel for dual fuel mode at the full load. Nevertheless, NO emission was slightly increased with modified fuel compared to F2. From these findings, it can be suggested that 60 ppm multi-walled carbon nanotubes additive can be an optimum level for combustion, performance, and emissions under the dual fuel mode.
  • [ X ]
    Öğe
    Dual-function macroalgae biochar: Catalyst for hydrogen production and electrocatalyst
    (Elsevier Sci Ltd, 2024) Bekirogullari, Mesut; Abut, Serdar; Duman, Fatih; Hansu, Tulin Avci
    In the current study, Enteromorpha intestinalis, a green macroalgae, has been utilized as a substrate to synthesise a new environmentally friendly and cheap dual -functional (catalyst and electrocatalyst) material. The catalyst was used for efficient hydrogen production from alcoholises sodium borohydride and as an anode catalyst for direct methanol fuel cell applications. At the catalyst synthesis stage, orthogonal arrays of Taguchi is used to find the optimum levels of independent variables for the superior catalyst performance that bears the modest kinetics. The experiments performed in accordance with the L16(45) type orthogonal array. The samples treated at relatively moderate acid, impregnation temperature, impregnation time, burning temperature and burning time showed higher catalytic activities with Exp(5) presenting the optimal catalytic activity followed by Exp(1), Exp (15), Exp(8) and Exp(12), respectively. The experimental levels of coded variables (acid molarity, impregnation temperature and time, burning temperature and time) for Exp(5) were 3 M, 50 C-degrees, 24 h, 500 C-degrees and 2 h., respectively. These finding suggest that providing maximum levels of each of independent variable will not provide high catalytic activity. Taking into account the binary and ternary interactions is an efficient way to determine optimal level of each parameter with regards to maximum hydrogen production rate. The morphological and structural characterization of the optimal catalyst was finally carried out with SEM- EDS, XRD and FT-IR. CV and EIS measurements were taken to determine the electrocatalytic activity of final biochar. Experimental studies revealed that the utilization of E. intestinalis-based electrocatalyst as an anode catalyst for direct fuel cell applications is promising. By employing advanced Taguchi -based experimentation, this research establishes the optimum levels of independent variables, enhancing the catalyst's performance, and highlighting a novel approach to modest kinetics improvement.
  • [ X ]
    Öğe
    Effect of stirring speeds on biodiesel yield using an innovative oscillatory reactor and conventional STR (A comparative study)
    (Elsevier Sci Ltd, 2022) Khelafi, Mostefa; Djaafri, Mohammed; Kalloum, Slimane; Atelge, M. R.; Abut, Serdar; Dahbi, Abdeldjalil; Bekirogullari, Mesut
    This paper aims to study the effect of stirring speed on biodiesel yield using an innovative oscillating reactor compared to the conventional stirring tank reactor. The efficiency of the invented reactor was compared with the conventional system, employing two catalysts (a homogeneous catalyst and a heterogeneous bio-catalyst). The obtained results showed that under low agitation speed of 50 rpm, the invented oscillating reactor is more efficient than the conventional system with a biodiesel yield of 93% compared to 90.13% using the heterogeneous catalyst and 93.53% compared to 92.7% using the homogeneous catalyst respectively. As for the higher stirring speeds, the conventional system was found to be slightly more efficient than the oscillating reactor when using the heterogeneous biocatalyst (96.03% against 94.42%) while the contrary was observed when using the homogeneous catalyst (94.43% against 95%). However, this slight increase in the biodiesel yield at higher speeds results in increasing production costs. This indicates that biodiesel production using the innovative oscillating reactor at low speeds is more economically viable. The characteristics of the produced biodiesel using the invented reactor were in agreement with the ASTM D6751 biodiesel standards. Moreover, a two-way ANOVA analysis was conducted to compare between groups that have been split on two independent variables as reactor type and stirring speed. The statistical analysis proved that the invented oscillating reactor performs better when using heterogeneous catalysts at low stirring speed levels. This study suggests that the biodiesel yield of the innovative reactor can be further enhanced by introducing a baffle system which provides a relatively larger contact surface area. Similarly, synthesis of other heterogeneous bio-catalysts derived from the date seed of another date palm cultivar can be tested to further improve the biodiesel yield.
  • Küçük Resim
    Öğe
    The effects of intravenous lipid emulsion therapy in the prevention of depressive effects of propofol on cardiovascular and respiratory systems: an experimental animal study
    (25.11.2018) Doğanay, Fatih; Ak, Rohat; Alışkan, Halil; Abut, Serdar; Sümer, Engin; Onur, Özge
    Background and objectives: Although there are several hypotheses about the mechanism of action, intravenous lipid emulsion (ILE) therapy has been shown to be effective in the treatment of toxicities due to local anaesthetics and many lipophilic drugs. In this study, we had hypothesized that ILE therapy might also be effective in preventing mortality and cardiorespiratory depressant effects due to propofol intoxication. Materials and methods: Twenty-eight Sprague-Dawley adult rats were randomly divided into four groups. Saline was administered to the subjects in the control group. The second group was administered propofol (PP group); the third group was administered ILE (ILE group), and the fourth group was administered propofol and ILE therapy together (ILE+PP group). Systolic blood pressure (SBP), diastolic blood pressure (DBP), respiratory rate (RR), heart rate (HR), and mortality were recorded at 10 time-points during a period of 60 min. A repeated measures linear mixed-effect model with unstructured covariance was used to compare the groups. Results: In the PP group; SBP, DBP, RR, and HR levels declined steadily; and all rats in this group died after the 60-min period. In the ILE+PP group, the initially reduced SBP, DBP, RR, and HR scores increased close to the levels observed in the control group. The SBP, DBP, RR, and HR values in the PP group were significantly lower compared to the other groups (p < 0.01). The mortality rate was 100% (with survival duration of 60 min) for the PP group; however, it was 0% for the remaining three groups. Conclusions: Our results suggest that the untoward effects of propofol including hypotension, bradycardia, and respiratory depression might be prevented with ILE therapy.
  • [ X ]
    Öğe
    Investigation of Dunaliella salina microalgae as an effective dual-function material for hydrogen production and supercapacitor applications
    (Pergamon-Elsevier Science Ltd, 2024) Cetin, Ridvan; Kaya, Mustafa; Akdemi, Murat; Arseri, Muhammet Ali; Abut, Serdar
    Today, population growth, industrialization and economic growth increase the con-sumption of fossil fuels to meet the energy demand. The scarcity of fossil fuels and the harmful gases they generate increase the interest in renewable energy sources. One of these sources is hydrogen energy, which is plentiful in nature and has no negative envi-ronmental effects. Sodium borohydride (NaBH4) is a good source of hydrogen, but a catalyst must used for methanolysis. Besides producing energy, it is also important to store it. Supercapacitors are a good alternative to energy storage elements due to their outstanding advantages. In this work, Dunaliella salina (DS) microalgae were used as substrate to syn-thesize activated carbon for the first time to develop materials that can operate both as a catalyst and an electrode material for supercapacitors. The activated carbon was obtained by carbonization and activation and the taguchi experimental approach was used to minimize the number of experiments. The best hydrogen production rate (HPR) result for DS-9 catalyst with 0.10 g catalyst and 0.25 g NaBH4 at ambient temperature of 60 degrees C was determined to be 13,085 mL min(-1)gcat(-1). The material with the best HPR value was then used as the electrode material for supercapacitor design. The specific capacitance value for 1 A/g was determined using galvanostatic charge-discharge (GCD) curves to be 216 F/g. In addition, the produced supercapacitor has an energy density of 13.80 W h/kg at a power density of 480 W/kg. The results indicate that the ecologically friendly and cost effective bifunctional materials produced can be used both in reuse of organic wastes and in catalyst and supercapacitor applications. (c) 2023 Hydrogen Energy Publications LLC. Published by Elsevier Ltd. All rights reserved.
  • [ X ]
    Öğe
    Lake sediment based catalyst for hydrogen generation via methanolysis of sodium borohydride: an optimization study with artificial neural network modelling
    (Springer, 2021) Bekirogullari, Mesut; Abut, Serdar; Duman, Fatih; Hansu, Tulin Avci
    In the current study, lake sediment, a heterogeneous and complex organic matter, utilized as a catalyst upon acid treatment for efficient hydrogen generation from sodium borohydride. In order to synthesise the catalyst that bears the best catalytic activity, ANOVA, cubic stepwise linear regression and artificial neural network optimization techniques were applied to determine the optimal level of treatment parameters. The results suggest that only Taguchi orthogonal arrays method was able to accurately reflect the overall surface of objective variable. Among the 16 catalyst samples Exp(15) showed the superior catalytic activity followed by Exp(13), Exp(12), Exp(14) and Exp(7). The minimum reaction completion time for Exp(15) corresponding to maximum hydrogen production rate of 3247.15 mL/min/gcat was 2.25 min. A detailed characterization of the final product was carried out by using a Fourier transform infrared spectra (FTIR-Perkin Elmer), an X-ray diffractometer (Bruker D8 Advance XRD), a scanning electron microscopy and energy dispersive X-ray spectroscopy. [GRAPHICS] .
  • [ X ]
    Öğe
    Modeling and simulation of co-digestion performance with artificial neural network for prediction of methane production from tea factory waste with co-substrate of spent tea waste
    (Elsevier Sci Ltd, 2021) Ozarslan, Saliha; Abut, Serdar; Atelge, M. R.; Kaya, M.; Unalan, S.
    The production of biofuel from waste has become an important topic for waste management and reducing its environmental hazard. Tea factory waste is a strong candidate due to its availability and sourceability. This study aimed to reveal the biochemical methane potential (BMP) of tea factory waste (TFW) and spent tea waste (STW). Additionally, the results revealed that both substrates had high biodegradability due to high VS removal. The BMP tests took 49 days under mesophilic conditions with a batch reactor and the cumulative methane yields were 249 +/- 3, and 261 +/- 8 mL CH4/g VS for TFW and STW, respectively. According to prediction data with the selected ANN model, which was 50 hidden layer sizes, trained with Bayesian Regularization algorithm, the maximum cumulative specific methane yield of the co-digestion was simulated as 468.43 mL CH4/g VS when the ratio of 65 and 35% (w/w by VS) of TFW and STW, respectively. The predicted methane yield for co-substrates was 183% higher than mono substrates. This result revealed that TFW can be a good candidate for biogas production as biofuel for not only its availability and sourceability but also the synergistic effect possible for codigestion.
  • Küçük Resim
    Öğe
    Novel approach to study dispersion in growth and dissolution rate of crystals from solutions
    (2019) Kaya, Mustafa; Ceyhan, Abdullah; Abut, Serdar; Şahin, Ömer
    Growth and dissolution rate dispersion of boric acid single crystal was investigated in supersaturated and undersaturated solution contained in a specially designed cell by in situ examination of crystals mounted at the tip of a needle, rotated between 0° and 360°, and using an image analyzer system. Changes in the behavior of equivalent crystal diameters at different supersaturation and undersaturation and rotation angles were determined by using two new procedures: (1) rotating growing or dissolving crystals by 360° at intervals of 45° (Procedure 1: angular growth/dissolution rates), and scanning individual face at increments of 10° in the 360° range (Procedure 2: point growth/dissolution rates). It was found that: (1) dispersion in angular growth and dissolution rates is a general phenomenon due to different growth and dissolution rates of individual faces of a crystal, but this dispersion is reduced when the average growth and dissolution rates of all faces are taken into account, and (2) the point growth/dissolution rates of single crystal, calculated by procedure 2, are different from point to point.
  • [ X ]
    Öğe
    Paradigm shift from Artificial Neural Networks (ANNs) to deep Convolutional Neural Networks (DCNNs) in the field of medical image processing
    (Pergamon-Elsevier Science Ltd, 2024) Abut, Serdar; Okut, Hayrettin; Kallail, K. James
    Images and other types of unstructural data in the medical domain are rapidly becoming data-intensive. Actionable insights from these complex data present new opportunities but also pose new challenges for classification or segmentation of unstructural data sources. Over the years, medical problems have been solved by combining traditional statistical methods with image processing methods. Both the increase in the size of the data and the increase in the resolution are among the factors that shape the ongoing improvements in artificial intelligence (AI), particularly concerning deep learning (DL) techniques for evaluation of these medical data to identify, classify, and quantify patterns for clinical needs. At this point, it is important to understand how Artificial Neural Networks (ANNs), which are an important milestone in interpreting big data, transform into Deep Convolutional Neural Networks (DCNNs) and to predict where the change will go. We aimed to explain the needs of these stages in medical image processing through the studies in the literature. At the same time, information is provided about the studies that lead to paradigm shift and try to solve the image related medical problems by using DCNNs. With the increase in the knowledge of medical doctors on this subject, it will be possible to look at the solution of new problems in computer science from different perspectives.
  • [ X ]
    Öğe
    The Importance of artificial neural networks in decision making for the field of medicine
    (Nova Science Publishers, Inc., 2024) Abut, Serdar; Okut, Hayrettin
    This book chapter explores the integration of machine learning techniques, particularly deep neural networks, in the field of medical image processing for precision medicine. The healthcare industry has accumulated vast amounts of complex data, and advancements in technology have led to an increase in structured and unstructured medical data. The chapter discusses the historical development of image processing techniques, moving from labor-intensive approaches to more efficient and faster operations using artificial neural networks. Various feature extraction methods, with a focus on dimensionality reduction, are investigated to optimize the performance of neural networks. The application of deep neural network models in medical imaging is explored, with a gradual implementation strategy proposed to address challenges related to data variability across institutions. The potential benefits and obstacles of using deep neural network models for psychiatric diagnoses and neonatal early detection are discussed. Throughout the chapter, the importance of effective communication between data scientists, software engineers, and clinicians is emphasized for the development of robust and practical artificial intelligence systems in healthcare. The overall viewpoint highlight how technological advancements in image processing have transformed artificial intelligence-based applications in the medical field and future developments can be looked at in a different light in this area. © 2024 by Nova Science Publishers, Inc. All rights reserved.
  • [ X ]
    Öğe
    Unsupervised learning in civil engineering
    (Nova Science Publishers, Inc., 2024) Abut, Yavuz; Abut, Serdar
    This chapter explores the application of unsupervised learning in civil engineering, focusing on its advantages and challenges. Unsupervised learning is a machine learning approach that is becoming increasingly popular in the field of civil engineering. This method utilizes the model's ability to learn from unlabeled datasets and focuses on uncovering structures and patterns within the data. This type of learning offers several benefits for civil engineers. One advantage of unsupervised learning methods is the ability to analyze large amounts of unlabeled data more effectively. Labeling datasets, especially in complex data types such as images or sensor data, can be a tedious and time-consuming task. Unsupervised learning provides a more efficient alternative to overcome this challenge. Another advantage is the capability to discover hidden structures and patterns within datasets, allowing for deeper analysis. For example, these methods can be utilized to detect early signs of deformation or damage in a structure. By identifying similarities and differences within the dataset, these methods can detect damaged areas or abnormal behavior. Furthermore, unsupervised learning methods can help civil engineers in discovering features within their datasets. This is particularly important in large datasets or those obtained from various sources. By extracting features from the dataset, unsupervised learning methods can improve data representation and yield better results. In conclusion, the application of unsupervised learning in civil engineering can enhance the data analysis and pattern discovery processes. These methods provide civil engineers with valuable insights by leveraging information from unlabeled datasets, thereby aiding in making better decisions. With the expected increase in unsupervised learning studies in civil engineering, we can anticipate more application areas and further advancements in techniques in the future. © 2024 by Nova Science Publishers, Inc. All rights reserved.