2022 - Publications Repository - Gdańsk University of Technology

Show publications from the year

Publications from the year 2022

Show all

Układ sterowania 5-fazową maszyną synchroniczną z magnesami trwałymi (PMSM)
- Krzysztof Łuksza
- Jarosław Guziński
- Marcin Morawiec
- Filip Wilczyński
2022 Automatyka Elektryka Zakłócenia
W pracy przedstawiono nowy układ sterowania bezczujnikowego 5-fazową maszyną synchroniczną z magnesami trwałymi zagłębionymi (IPMSM) . Zaproponowany układ wykorzystuje obserwator stanu dla zmiennych w pierwszym układzie współrzędnych (1 harmoniczna), jak i drugim (3 harmoniczna), w celu zwiększenia gęstości momentu elektromagnetycznego. Zaprezentowano schemat kompletnego układu sterowania, przebiegi zmiennych w stanie ustalonym, jak również podczas stanów przejściowych, zarejestrowane podczas badań laboratoryjnych z wykorzystaniem prototypowej 5-fazowej maszyny PMSM.
Ultrasound-assisted deep eutectic solvent-based liquid–liquid microextraction for simultaneous determination of Ni (II) and Zn (II) in food samples
- Fazal Elahi
- Muhammad Balal Arain
- Wajid Ali Khan
- Hameed Haq
- Asif Khan
- Faheem Jan
- Roberto Castro-Muñoz
- Grzegorz Boczkaj
2022 Full text FOOD CHEMISTRY
A new approach was developed for the simultaneous pre-concentration and determination of Ni (II) and Zn (II) in food samples. This method is based on ultrasound-assisted liquid–liquid micro extraction using hydrophobic deep eutectic solvent (DES) and 1,10-phenanthroline as chelating agent. The effect of several parameters, such as pH, selection and volume of DES, amount of chelating agent, time of sonication and centrifugation, was studied. Under optimized conditions, the developed procedure offered exceptional sensitivity and linearity. The limit of detection was approximately 0.029 µg/Kg and 1.5 µg/Kg for Ni (II) and Zn (II), respectively. The proposed method was applied for the pre-concentration and determination of Ni (II) and Zn (II) in hydrogenated edible oils, fishes, and milk samples. The results of this study were compared with reported methods in the literature revealing its advantages.
Ultrasound-Assisted Dispersive Liquid-Liquid Microextraction Using Deep Eutectic Solvents (DESs) for Neutral Red Dye Spectrophotometric Determination
- Sana Ullah
- Hameed Haq
- Muhammad Salman
- Faheem Jan
- Faisal Safi
- Muhammad Balal Arain
- Muhammad Khan
- Roberto Castro-Muñoz
- Grzegorz Boczkaj
2022 Full text MOLECULES
Deep eutectic solvents (DES), which have low toxicity and are low cost, biodegradable, and easily synthesized, were used for the extraction of neutral red (NR) dye before its spectrophotometric analysis. DES, containing choline chloride as a hydrogen bond acceptor and phenol as a hydrogen bond donor with a molar ratio of 1:2, was used for the extraction of NR dye from aqueous media. The possible interaction of different DESs with NR was studied using density functional theory (DFT) calculations. Experimentally, a UV-visible spectrophotometer was used for the quantitative analysis. The most important parameters affecting method performance, such as pH, extraction temperature, DES type, its volume, THF volume, sonication time, and centrifugation time, were optimized. The developed method provides exceptional sensitivity in terms of LOD and LOQ, which were 2.2 and 7.3 µg/L respectively. The relative standard deviation was 1.35–1.5% (n = 10), and the pre-concentration factor was 40. The method was found to be linear in the range of 2–300 µg/L (R2 = 0.9967). The method was successfully used for the determination of NR in wastewater samples. Finally, the DES-based method presents operational simplicity, high sensitivity, and rapid determination (<5 min) compared with other analytical procedures.
Uncertainty quantification of modal parameter estimates obtained from subspace identification: An experimental validation on a laboratory test of a large-scale wind turbine blade
- Szymon Greś
- Riccardo Riva
- Cem Yeniceli Süleyman
- Palle Andersen
- Marcin Łuczak
2022 Full text ENGINEERING STRUCTURES
The uncertainty afflicting modal parameter estimates stems from e.g., the finite data length, unknown, or partly measured inputs and the choice of the identification algorithm. Quantification of the related errors with the statistical Delta method is a recent tool, useful in many modern modal analysis applications e.g., damage diagnosis, reliability analysis, model calibration. In this paper, the Delta method-based uncertainty quantification methodology is validated for obtaining the uncertainty of the modal parameter and the modal indicator estimates in the context of several well-known subspace identification algorithms. The focus of this study is to validate the quality of each Delta method-based approximation with respect to the experimental Monte Carlo distributions of parameter estimates using a statistical distance measure. On top of that, the accuracy in obtaining the related confidence intervals is empirically assessed. The case study is based on data obtained from an extensive experimental campaign of a large scale wind turbine blade tested in a laboratory environment. The results confirm that the Delta method is, on average, adequate to characterize the distribution of the considered estimates solely based on the quantities obtained from one data set, validating the use of this statistical framework for uncertainty quantification in practice.
Unconventional and user-friendly sampling techniques of semi-volatile organic compounds present in an indoor environment: An approach to human exposure assessment
- Mariusz Marć
- Bożena Zabiegała
2022 Full text TRAC-TRENDS IN ANALYTICAL CHEMISTRY
The commonly applied solutions used to assess the potential risk of human exposure to semi-volatile organic compounds (SVOCs) are based on the investigation of biological samples collected in an invasive or non-invasive manner. For SVOCs, which are typically introduced to humans through the respiratory system, dermal adsorption, or digestive system, sampling solutions generally used in the indoor environments are classified as active and passive. From the user's perspective, the most convenient method to assess the potential risk is the use of an analytical tool that combines the benefits of passive and non-invasive sampling techniques—use of an unconventional personal sampler such as a silicone wristband, brooch, dog tag, cotton gauze, or viscose wiper. Despite the advantages of this method, the aforementioned techniques require further analytical research owing to the differences in the results of human exposure assessment owing to the lack of standards and unified sampling protocols.
Uncovering the invisible barriers to women’s success
- Agnieszka Lendzion
2022
In the area of science and higher education, as in business and politics, the job situation of women is improving, and the percentage of women on executive positions is increasing. However, there is still a serious underrepresentation of women in the highest decision-making bodies. Ladies also take part in the strategic institutional events less frequently. There are still serious disproportions in academic and management positions, and the research career of women in relation to men slows down and encounters various barriers. Universities and research institutions undertake various bottom-up and top-down, voluntary and forced gender equality initiatives. Nevertheless, women still face the phenomena of the glass ceiling, glass escalator, discrimination, violent behavior, income inequality and other barriers on their career paths. What makes barriers and stereotypes still present in a well-educated, aware academic and research environment? What can we do to overcome these problems? How to introduce changes effectively so that women climb (not only the scientific) career levels with a similar and the same impetus as men, so that research teams and decision-making bodies become diverse? During the discussion panel dedicated to uncover the invisible barriers to women’s success, we will try to answer these questions, share good practices and discuss potential solutions supporting women in academic environment.
Understanding ion–ion and ion–solvent interactions in aqueous solutions of morpholinium ionic liquids with N-acetyl-L-alaninate anion through partial molar properties and molecular dynamics simulations
- Dorota Warmińska
- Maciej Śmiechowski
2022 JOURNAL OF MOLECULAR LIQUIDS
Amino acid ionic liquids (AAILs) provide a low toxicity, biodegradable alternative to conventional ionic liquids, while also maintaining solubility in water. Densities and sound velocities of aqueous solutions of four amino acid ionic liquids (AAILs), based on the N-alkyl-N-methylmorpholinium ([Mor1,R], R = 2, 3, 6, 8) cation and N-acetyl-L-alaninate ([N-Ac-L-Ala]) anion were measured at T = (293.15–313.15) K and at atmospheric pressure. These data were used to derive the apparent molar volumes and the apparent molar compressibilities in the concentration range of (0.02–0.3) mol kg−1. Established extrapolation procedures were applied to estimate the values of the limiting apparent molar volumes and the limiting apparent molar compressibilities. The limiting apparent molar expansibility coefficients were obtained and the hydration numbers for AAILs in aqueous solution were calculated using Passynski’s method. The results indicate that ion–solvent interactions in the studied solutions are determined by the hydrophobic effect and the charge effect, which are both dependent on temperature. At low temperatures the ion–solvent interactions are determined by the hydrophobic effect which results in the decrease of the limiting apparent molar compressibilities with an elongation of the alkyl chain of the cation of AAIL. At high temperatures the charge effect is dominant due to the decreasing limiting apparent molar compressibilities. On the other hand, ion–ion interactions are dominated by the hydrophobic effect regardless of the temperature. Molecular dynamics simulations were also performed and applied in interpretation of experimental data at atomistic detail. Computed limiting apparent molar volumes correlate well with the values derived from experiment. The apparent hydration volume, obtained by subtracting the intrinsic volume contribution, is large and positive, indicating hydrophobic hydration and not electrostriction is the principal mode of ions’ influence on the solvent. Analysis of radial distribution functions confirms insignificant share of ion–ion interactions via contact ion pairing. The solvation shells signify predominantly hydrophobic hydration, with locally strongly hydrated functional groups.
Understanding the early-stage release of volatile organic compounds from rapeseed oil during deep-frying of tubers by targeted and omics-inspired approaches using PTR-MS and gas chromatography
- Tomasz Majchrzak
- Mariusz Marć
- Andrzej Wasik
2022 Full text FOOD RESEARCH INTERNATIONAL
During deep-frying, a plethora of volatile products is emitted with the fumes. These compounds could act as oil quality indicators and change the indoor air composition leading to health risks for occupants. The presented experiments focus on deep-frying of different tubers in rapeseed oil at different frying temperatures. Here, two scenarios for real-time monitoring of volatile organic compounds (VOCs) using proton transfer reaction mass spectrometry (PTR-MS) were proposed. The first, targeted, involved the application of gas chromatography with a flame ionization detector (GC-FID). The second, omics-inspired, involved the use of solid-phase microextraction (SPME) along with gas chromatography-mass spectrometry (GC-MS) and molecular networking algorithm as a complementary tool to the PTR-MS analysis. In a targeted approach, it was shown that the emission profile of pentanal and hexanal depends on the frying temperature and as the temperature increases, a sudden release of these compounds can be observed in the first minutes of frying. Meanwhile, using an omics-inspired protocol enables finding the relation between 1,4-heptadienal and 2-heptanone, octanal and limonene emissions. Using both approaches it was possible to record real-time changes in emission profiles of various oils’ degradation products. It was also observed that the emission profiles of VOCs are strictly related to the frying temperature and the type of fried tuber.
Understanding the Electronic Structure and Optical Properties of Vacancy-Ordered Double Perovskite A2BX6 for Optoelectronic Applications
- Muhammad Faizan
- Xinjiang Wang
- Shaimaa A. M. Abdelmohsen
- K. C. Bhamu
- Subrahmanyam Sappati
- Amel Laref
- Nisar Muhammad
- Muhammad Mushtaq
- Ashraf M. M. Abdelbacki
- Rabah Khenata
2022 Full text ENERGY & FUELS
Over the past few years, metal halide perovskite solar cells have made significant advances. Currently, the single-junction perovskite solar cells reach a conversion efficiency of 25.7%. Perovskite solar cells with a wide band gap can also be used as top absorber layers in multi-junction tandem solar cells. We examined the dynamical and thermal stability, electronic structure, and optical features of In2PtX 6 (X = Cl, Br, and I) perovskites, utilizing first-principle calculations. The stability is predicted using phonon dispersion spectrum and ab initio molecular dynamics simulation and also through the convex hull approach. The lattice constants and the optimized volume show an increasing trend with changing halide ions. The band structures computed for In2 PtCl6, In 2PtBr 6, and In2 PtI6 indicate their semiconducting nature with band gap values of 2.06, 2.01, and 1.35 eV, respectively. Halogens p and Pt d orbitals, respectively, play a prominent role in the formation of states around valence band maximum and conduction band minimum. The compounds, namely, In2PtBr 6 and In2PtI 6, exhibit high dielectric constants and small carrier effective masses. Furthermore, we found that In2 PtI 6 reveals a maximum theoretical efficiency owing to its optimum band gap and high optical absorption and is comparable to MAPbI3 in the studied range. Our results suggest that In 2PtX 6 (X = Cl, Br, and I) are suitable materials for single-junction and top absorber layers in tandem solar cells.
Underwater Processing of Materials
- Dariusz Fydrych
- Jacek Tomków
2022 Full text Materials
Editorial do SI
Underwater wet welding of S1300 ultra-high strength steel
- Jacek Tomków
- Michał Landowski
- Dariusz Fydrych
- Grzegorz Rogalski
2022 Full text MARINE STRUCTURES
Marine Structures Volume 81, January 2022, 103120 Underwater wet welding of S1300 ultra-high strength steel Author links open overlay panelJacekTomkówGrzegorzRogalski https://doi.org/10.1016/j.marstruc.2021.103120 Get rights and content Under a Creative Commons licenseopen access Highlights • Technological method of S1300 steel underwater weldability improving was proposed. • Number of cracks and hardness of welded joints was reduced by Temper Bead Welding (TBW). • Thickness of the base metal does not affect the effectiveness of the TBW technique. • The highest TBW effectiveness was found for 50% beads overlap. Abstract An expected tendency in marine and offshore structures is an increasing share of steels with high mechanical properties. The necessity to perform the welding work under water leads to a decreasing of the weldability of the steel. It results from the increased susceptibility to form cold cracks, especially in welded structures made of high-strength steel. A potential method, which may improve the problem of limited weldability of steel under water is the use of temper bead welding (TBW) technique. In this paper, the possibility of underwater wet welding by covered electrode S1300 ultra-high strength steel of different thickness (4 mm and 10 mm) has been investigated. The visual, metallographic macro- and microscopic tests, and Vickers HV10 measurements of samples with two nonparallel beads were performed. On the basis of the performed tests, the effectiveness of TBW technique was quantitatively verified. It was proved that TBW leads to positive microstructural changes and reduces the number of cracks in different regions of welded joints. Moreover, it was found that the proposed technique allows to decrease the hardness of S1300 steel joint by 30–80 HV10 in the weld metal and 40–100 HV10 in heat-affected zone. Finally, it was stated that the thickness of the base material does not affect the effectiveness of the TBW technique.
Universal construction of genuinely entangled subspaces of any size
- Maciej Demianowicz
2022 Full text Quantum Journal
We put forward a simple construction of genuinely entangled subspaces – subspaces supporting only genuinely multipartite entangled states – of any permissible dimensionality for any number of parties and local dimensions. The method uses nonorthogonal product bases, which are built from totally nonsingular matrices with a certain structure. We give an explicit basis for the constructed subspaces. An immediate consequence of our result is the possibility of constructing in the general multiparty scenario genuinely multiparty entangled mixed states with ranks up to the maximal dimension of a genuinely entangled subspace.
Unusual divergence of magnetoacoustic beams
- Anna Perelomova
2022 Full text PHYSICS OF PLASMAS
Two-dimensional magnetosonic beams directed along a line forming a constant angle h with the equilibrium straight magnetic field are considered. Perturbations in a plasma are described by the system of ideal magnetohydrodynamic equations. The dynamics of perturbations in a beam are different in the cases of fast and slow modes, and it is determined by h and equilibrium parameters of a plasma. In particular, a beam divergence may be unusual in the case of parallel propagation (h ¼ 0). Diffraction is more pronounced in the case of parallel propagation as compared to a flow without magnetic field, and less manifested in the case of perpendicular propagation. The beams propagating oblique to the magnetic field do not reveal diffraction. The dynamics of perturbations in a beam are analytically described in the cases of weak and strong nonlinearity compared to diffraction. Small magnitude perturbations at the axis of a beam in unusual cases propagate slower than that in the plane wave. Involving of thermal conduction leads to the coupling equations describing thermal self-action of a beam, which behaves differently in the ordinary and unusual cases. Self-focusing may occur in the presence of a magnetic field instead of conventional defocusing in gases
Up-to-date strategies and future trends towards the extraction and purification of Capsaicin: A comprehensive review
- Roberto Castro-Muñoz
- Emilia Gontarek-Castro
- Seid Mahdi Jafari
2022 Full text TRENDS IN FOOD SCIENCE & TECHNOLOGY
Background According to the current need of manufacturing healthier products, food companies are seeking specific biomolecules that may offer additional added value (i.e., biological activities) to the new food formulations. Capsaicin, as the pungent ingredient of chili peppers, has become so far one of the target biomolecules explored since the 1950s. There is evidence demonstrating that capsaicin exhibits important biological properties in human health including inhibits acid secretion, stimulates alkali and mucus secretion and particularly gastric mucosal blood flow contributing to the prevention and healing of gastric ulcers, thermoregulation, among many other reported bioactivities. Scope and approach However, one of the main bottlenecks deals with the proper protocol of extraction and purification of this compound since most of the conventional methods based on solvent extraction do not provide efficient yield, along with diminished bioactivity of the compounds. Therefore, this review comprehensively elucidates the current strategies proposed by researchers towards the sustainable extraction and purification of capsaicin from its natural source, and comparison over traditional extraction methods. Particular emphasis has been focused on the innovative extraction techniques and the relevant insights over the last five years. Key findings and conclusion A detailed discussion is provided on the advantages and drawbacks of the novel techniques, key interactions with target molecules and their effect on the bioactivity of capsaicin. To finalize, according to the findings of this review, the future trends, perspectives, and research gaps are also given.
Urinary Tract Infections Caused by K. pneumoniae in Kidney Transplant Recipients – Epidemiology, Virulence and Antibiotic Resistance
- Beata Krawczyk
- Magdalena Wysocka
- Michał Michalik
- Justyna Gołębiewska
2022 Full text FRONTIERS IN CELLULAR AND INFECTION MICROBIOLOGY
Urinary tract infections are the most common complication in kidney transplant recipients, possibly resulting in the deterioration of a long-term kidney allograft function and an increased risk of recipient’s death. K. pneumoniae has emerged as one of the most prevalent etiologic agents in the context of recurrent urinary tract infections, especially with multidrug resistant strains. This paper discusses the epidemiology and risk factors associated with urinary tract infections in kidney transplant recipients, multi-drug resistance of K. pneumoniae (ESBL, KPC, NDM), treatment and pathogenesis of K. pneumoniae infections, and possible causes of recurrent UTIs. It also addresses the issue of colonization/becoming a carrier of K. pneumoniae in the gastrointestinal tract and asymptomatic bacteriuria in relation to a symptomatic UTI development and epidemiology.
Use of a wastewater recovery product (struvite) to enhance subtropical seagrass restoration
- Conor Macdonnell
- Franciszek Bydałek
- T. Z. Osborne
- A. Beard
- S. Barbour
- D. Leonard
- Jacek Mąkinia
- P.w. Inglett
2022 Full text SCIENCE OF THE TOTAL ENVIRONMENT
Seagrasses are in decline worldwide, and their restoration is relatively expensive and unsuccessful compared to other coastal systems. Fertilization can improve seagrass growth in restoration but can also release nutrients and pollute the surrounding ecosystem. A slow-release fertilizer may reduce excessive nutrient discharge while still providing resources to the seagrass's rhizosphere. In this study, struvite (magnesium ammonium phosphate), a relatively insoluble, sustainable compound harvested in wastewater treatment plants, was compared to Osmocote™ (14:14:14 Nitrogen: Phosphorus: Potassium, N:P:K), a popular polymer coated controlled release fertilizer commonly used in seagrass restoration. Two experiments compared the effectiveness of both fertilizers in a subtropical flow-through mesocosm setup. In the first experiment, single 0.5 mg of P per g dry weight (DW) doses of Osmocote™ and struvite fertilizers were added to seagrass plots. Seagrass shoot counts were significantly higher in plots fertilized with struvite than both the Osmocote™ and unfertilized controls (p < 0.0001). A significant difference in total P concentrations was observed in porewater samples of Osmocote™ vs struvite and controls (p < 0.0001), with struvite fertilized plots emitting more than controls (p ≤ 0.0001), but less than 2% of the total dissolved P (TDP) of Osmocote™ fertilized plots (100+ mg/L versus x > 5 mg/L). A subsequent experiment, using smaller doses (0.01 and 0.025 mg of P per gram DW added), also found that the struvite treatments performed better than Osmocote™, with 16–114% more aboveground biomass (10–60% higher total biomass) while releasing less N and P. These results indicate the relatively rapid dissolution of Osmocote™ may pose problems to restoration efforts, especially in concentrated doses and possibly leading to seagrass stress. In contrast, struvite may function as a slow-release fertilizer applicable in seagrass and other coastal restoration efforts.
Use of structural equation modeling in quantitative research in the field of management and economics: A bibliometric analysis in the systematic literature review
- Agnieszka Zakrzewska-Bielawska
- Anna Lis
- Anna Ujwary-Gil
2022 Full text Journal of Entrepreneurship, Management and Innovation
PURPOSE: This paper aims to provide a comprehensive review of scholarly research focusing on using quantitative methods and particularly structural equation modeling (SEM) in management and economics studies, as well as provide a bibliometric agenda including the time horizon of individual publications, the highest citation rate, geographic and industry areas, methodological context, and keywords. METHODOLOGY: A systematic literature review (SLR) was undertaken using the Web of Science and Scopus databases. We limited our search to the last five years to identify the newest research publications, and we used keywords related to quantitative research while excluding qualitative research. Then we analyzed papers related to SEM and those published in English. FINDINGS: Our results confirmed that quantitative methods are used both in management and economics research, and showed a growing trend in the number of publications in the last five years. However, there are many more publications on management than on economics as well as there are more papers published in the Scopus database than Web of Science. Taking into account structural equation modeling, this method is used primarily in management research. In terms of industry, publications using SEM considered both single- and multi-industry including, first, all Asian countries and then African ones. Publications, especially in the management field, are descriptive in nature and based on primary data collected using a survey questionnaire. Papers are published in various journals and the most cited are those published in journals with wider subject areas. IMPLICATIONS: The systematic literature review is a fundamental necessity in any field of knowledge, benefiting both academia and learners. Our results may be useful for future researchers planning research using quantitative methods, especially SEM, in the business or economic field, by indicating the most cited papers and journals as well as industry and country areas. ORIGINALITY AND VALUE: This paper represents a systematic attempt to link quantitative methods, with a particular emphasis on SEM, with research interests on managerial and economic subjects and papers published in the Web of Science and Scopus databases. Employing the bibliometric analysis within the systematic literature review, the paper shows interest and the current state of research using quantitative methods which proves its value and originality.
User Mobility’s Influence on System Loss in Off-Body BAN Scenarios
- Manuel M. Ferreira
- Filipe D. Cardoso
- Luis M. Correia
- Sławomir Ambroziak
- Kenan Turbić
2022
In this paper, a measurement campaign for off-body communications in an indoor environment is investigated for a set of on-body antennas. The channel impulse response was measured with the user approaching and departing from an off-body fixed antenna using two user dynamics: standing at fixed positions and walking. The processing of the measurement data allowed to evaluate system loss statistics. Different antenna configurations are classified in terms of mobility and visibility depending on on-body antenna placements. A dependence on distance is found for the antennas with the lowest mobility (chest and head), while no significant dependence is found for the antennas with highest mobility (arm and leg). Regarding the standard deviation for system loss, higher values are found in the walking scenarios (above 2.6 dB) compared to the standing ones (below 0.6 dB). Standard deviation also shows a clear dependence on mobility: 1.6 dB for walking and 0.5 dB for standing scenarios.
User-assisted methodology targeted for building structure interpretable QSPR models for boosting CO2 capture with ionic liquids
- Joachim Eichenlaub
- Paulina Rakowska
- Adam Kloskowski
2022 JOURNAL OF MOLECULAR LIQUIDS
Task-specific ionic liquid (IL) is an emerging class of compounds that may be environmentally friendly. Properly selected, these compounds may be green alternative to amine solutions and can replace them in post-combustion carbon dioxide (CO2) capture processes on an industrial scale. However, owing to the vast diversity of ions and their possible combinations, laboratory research is time consuming and expensive. Therefore, computational methods are preferred for assessing their potential applications. In this study, three quantitative structure–property relationship models based on six distinct descriptors were created to predict Henry’s law constant (HLC) of CO2 in 62 ILs. The statistical parameters of multiple linear regression, logistic regression, and partial least squares models were satisfactory. In all cases, the coefficients of determination (R2) exceeded 0,90, and both external and internal validation proved them to be reliable and predictive with Q2 and Rpred2 values exceeding 0,90 and 0,87, respectively. Three of the descriptors were attributed to cations, and three were attributed to anions. In contrast to many other models, the descriptors were chosen in a manner that ensured their interpretability. Each of the six descriptors was analyzed for its influence on HLC. On this basis, guidelines for designing the structure of ionic liquids with increased CO2 absorption capacity were developed.
Using Convolutional Neural Networks for Corneal Arcus Detection Towards Familial Hypercholesterolemia Screening
- Tomasz Kocejko
- Jacek Rumiński
- Magdalena Mazur-Milecka
- Marzena Romanowska-Kocejko
- Krzysztof Chlebus
- Jo Kang-Hyun
2022 Full text Journal of King Saud University-Computer and Information Sciences
Familial hypercholesterolemia (FH) is a highly undiagnosed disease. Among FH patients, the onset of premature coronary artery disease is 13 times higher than in the general population. Early diagnosis and treatment is essential to prevent cardiovascular diseases and their complications, and to prolong life. One of the clinical criteria of FH is the occurrence of a corneal arcus (CA) among patients, especially those under 45 years old. Therefore, by detecting a CA, it might be possible to reduce the number of undiagnosed FH cases. In this paper, we propose using convolutional neural networks (CNN) for automatic recognition of the presence of a corneal arcus. To achieve this goal, we created a dataset of images of irises containing different stages of CA as well as irises without a CA. The core of the dataset consists of images acquired from patients with a corneal arcus, enroled in the National Centre of Familial Hypercholesterolemia in Gdansk. To increase the number of images, the dataset was complemented with images downloaded from the Internet. This dataset created for training and testing the model consisted of nearly 4000 images. To detect a CA in photographic images, we tested neural network models based on the VGG16, ResNet and Inception architectures. Finally, the performance of the models was evaluated on a set of images acquired from volunteers with a custom mobile application. The accuracy of CA detection in a real life scenario was 88% and the F1 score was 86%