Yeliz Durgun

The Multidrug-resistant (MDR) Acinetobacter baumannii poses a critical threat in clinical settings due to its high environmental stability and resistance to conventional antibiotics. Antimicrobial blue light, particularly at 405 nm, has emerged as a promising non-invasive decontamination strategy. Objective: This study aimed to evaluate the photo-inactivation efficacy and determine the killing kinetics of 405 nm light against A. baumannii. Methods: Bacterial samples were exposed to a 405 nm light source for up to 120 minutes. Survival rates were compared with a dark control group, and the data were modeled using first-order Chick-Watson kinetics. Results: A significant time-dependent reduction in bacterial viability was observed. After 120 minutes of exposure, 405 nm light achieved a 98.6% inactivation success compared to the control group (p < 0.01). Kinetic analysis revealed an inactivation rate constant (k) of 0.0291 min-1 with a high coefficient of determination (R2 = 0.8888). The calculated half-life (t1/2) and decimal reduction time (D-value) were 23.84 and 79.19 minutes, respectively.These results demonstrate that 405 nm irradiation provides a potent and mathematically predictable bactericidal effect against MDR A. baumannii. The findings suggest that 405 nm light therapy serves as a viable alternative for the environmental decontamination of highly resistant pathogens.

Food adulteration in meat-based broths remains a major concern for consumer safety, requiring rapid and reliable analytical approaches. This study presents a low-cost and real-time detection system for identifying chicken broth adulteration in beef broth using VIS–NIR spectroscopy integrated with edge computing. Beef and chicken carcasses were processed under controlled conditions, and a total of 60 samples were prepared to obtain four broth mixtures: 0%, 25%, 50%, and 100% chicken broth (n = 15 per group). Spectral measurements were acquired using the AS7265x smart spectral sensor, covering 18 wavelengths between 410–940 nm, positioned at a fixed 3 cm distance from the sample under controlled illumination. The Arduino Nano 33 BLE processed spectral data locally (edge computing) and executed a real-time decision algorithm based on wavelength-specific intensity changes.Significant spectral differences were observed between the four mixture groups, particularly at 560 nm and 585 nm. One-way ANOVA confirmed highly significant effects of adulteration level on spectral intensity (p < 0.001), and Tukey HSD revealed clear pairwise separation between all groups. Confidence interval analysis further demonstrated that 560 nm and 585 nm provided the most discriminative response ranges. The system achieved consistent detection performance across 40 experimental trials, demonstrating high reliability.The proposed approach offers a rapid, non-destructive, and cost-effective method for broth authentication while enabling on-device, real-time decision-making. These findings highlight the potential of combining VIS–NIR sensing with edge computing for practical and scalable food fraud detection applications.

Climate change, pathogen-induced environmental impacts, pesticide use, habitat loss, and industrial agriculture are leading to a significant decline in biological diversity. These disruptions adversely affect ecological balance and have resulted in the decline of wild pollinators and honeybees, which play a vital role in natural ecosystems. Consequently, beekeeping has also come under threat, as honeybees are among the most important pollinator insects in nature. With advancements in technology, wireless sensor networks now widely used across various domains have also begun to be implemented for remote monitoring of beehives. Through these remote monitoring systems, the continuous observation of hive conditions has become possible.This study aims to develop a hive monitoring system that enables the real-time, remote observation of honeybee colonies via a wireless sensor network in order to prevent colony losses and enhance honey production. The proposed system consists of locally placed sensors positioned at different points within the hive to measure temperature, relative humidity, and comb weight, as well as a cloud server where all collected data are stored. In addition, the system incorporates software and a web-based platform capable of performing data analysis and enabling operations on the database for real-time monitoring of hive parameters. Users of the developed web system can access and analyze the collected data. Real-time monitoring of temperature and humidity inside the hive will provide beekeepers with valuable insights into the health and environmental conditions of their colonies. Ultimately, this system will support beekeepers, the apiculture industry, and the agricultural sector by contributing to emerging apicultural technologies and addressing the urgent need to mitigate honeybee colony collapse. The system was operated with an average data sampling interval of 2 seconds over a 48-hour test period, during which hive temperature was maintained within ±0.4 °C and humidity within ±3 % RH. These quantitative findings demonstrate the effectiveness of the proposed system in achieving stable real-time monitoring and control.

Arı kolonilerinin sağlık ve faaliyetlerine ilişkin sürekli ve doğru bilgi elde etmek, arı popülasyonlarının korunması ve sürdürülebilir tarım uygulamalarının desteklenmesi için hayati öneme sahiptir. Çalışmamız, çevresel ve akustik sensör verilerini bütünleştirerek arı kolonilerinin durumunu etkin şekilde izlemeyi hedeflemektedir. Özellikle çalışmada CO2 konsantrasyonu, TVOC konsantrasyonu, sıcaklık ve nem gibi çevresel parametrelerin yanı sıra akustik verilerin analizi de yer almaktadır. Kullanılan yöntem, çeşitli çevresel sensörlerden elde edilen verilerin zaman serisi analizi ve ses dosyalarından spektral özelliklerin çıkarılmasını içermektedir. Analiz sürecinde, normal ve özel koşulları ayırt etmek için aykırı değer analizi uygulanmıştır. Sonuçlarımız, çevresel parametrelerde zamanla önemli değişiklikler olduğunu göstermektedir. Özellikle, özel koşullarda CO2 ve TVOC konsantrasyonlarında önemli artışlar gözlemlenmiştir (CO2: 1200 ppm'den 1450 ppm'ye, TVOC: 0 ppb'den 450 ppb'ye). Ayrıca, ses analizi, iki koşul arasında belirgin spektral özellik farkları göstermiştir. Sonuç olarak, araştırmamız, çevresel ve akustik sensör verilerinin birleşiminin arı kolonilerinin durumunu kapsamlı ve doğru izlememize olanak sağladığını doğrulamaktadır. Bulgular, bu tür bir yaklaşımın izleme uygulamaları için değerli bir araç olabileceğini önermektedir.

Bu çalışma, trafik ortamlarında ambulans sirenlerinin tespiti için geliştirilen ve Polinom Destek Vektör Makinesi (SVM) algoritmasını temel alan bir ses sınıflandırma modelini sunmaktadır. Kenar bilişim teknolojilerini kullanarak gerçekleştirilen örnek toplama deneyleriyle desteklenen bu çalışma, gerçek zamanlı veri işleme ve gömülü sistemlerde kullanılmak üzere tasarlanmıştır. Model, trafik sesleri ve ambulans sirenleri arasında etkili bir ayrım yapabilme kapasitesine sahiptir. UMAP ve PCA analizleri, modelin yüksek boyutlu verileri düşük boyutlu uzaylarda başarılı bir şekilde işleyebildiğini ve farklı ses sınıflarını net bir şekilde ayırt edebildiğini göstermektedir. Confusion Matrix, çapraz doğrulama sonuçları ve öğrenme eğrisi, modelin hem eğitim hem de doğrulama setleri üzerinde yüksek doğruluk oranlarına ulaştığını ve tutarlı bir performans sergilediğini belirtmektedir. ROC Eğrisi ve F1 Skoru, modelin genel sınıflandırma başarısının yüksek olduğunu gösterirken, düşük bellek ve işlemci gereksinimleri modelin gömülü sistemlerde ve gerçek zamanlı uygulamalarda etkin bir şekilde çalışabileceğinin altını çizmektedir. Bu çalışmanın sonuçları, ambulans sirenlerinin tespiti ve genel trafik seslerinin sınıflandırılması alanında, kenar bilişim tabanlı gömülü sistemlerin ve gerçek zamanlı veri işlemenin önemli bir adım olduğunu göstermektedir. Modelin daha da geliştirilmesi ve çeşitli uygulama senaryolarına adapte edilmesi için gelecekteki çalışmalar büyük önem taşımaktadır.

Foodborne pathogens remain a major public health concern throughout the farm-to-fork chain. Although thermal decontamination methods are effective, they may compromise product quality in certain foods and increase energy demand, creating the need for alternative technologies that operate with minimal heat load. In this context, 405 nm visible light delivered by violet–blue LEDs has emerged as a promising non-chemical approach for microbial inactivation. Exposure at 405 nm can stimulate endogenous chromophores (e.g., porphyrins), promoting the generation of reactive oxygen species and inducing oxidative damage to critical cellular components. The present study investigates the time dependent photoinactivation response of Listeria monocytogenes under 405 nm LED illumination and discusses its potential applicability for food safety and hygiene practicesExperimental results revealed that a 120-minute exposure to 405 nm LED light achieved a maximum statistically significant reduction of 0.39 log10 CFU/mL (p 0.001) in the Listeria monocytogenes population compared to the non illuminated control group

Near-UV, Visible Light, and Near-Infrared Radiation are some of the light wavelength regions that are of great interest for energy, agriculture, medical, and food industry research. The use of light spectrum for non-destructive, real-time sensing for imaging, food quality, and safety evaluation is becoming increasingly important in all these fields. Edge computing, which enables real-time monitoring of devices and control using machine learning methods, is necessary to improve system stability, minimize errors and develop tools that facilitate robotic intervention. In this study, the effect and performance of a drying system using Near-UV-vis-NIR radiation measurement in food drying using edge computing is presented. The system comprises three multi-spectral sensors that allow 18 different measurements. Sensors are also placed to measure the weight of the objects, temperature, and humidity inside the cabin. The data acquired is processed in real-time using a microcontroller (Arduino Nano 33 BLE) that can perform machine learning algorithms and control the cabin. Edge computing enables data processing and analytic operations to be performed on the device, thus providing real-time results and control operations. In this study, the change in radiation levels and the effect on drying quality during the drying process of apple slices are investigated. The results show that measurements performed using edge computing technology can effectively be performed during the drying process of apple slices.

This study aims to detect starch adulteration in dairy products utilizing an artificial neural network (ANN) model. Globally, milk fraud represents a significant challenge to food safety, posing substantial health risks to consumers. In this context, spectral data derived from milk samples with varying starch concentrations were processed using feature scaling and normalization techniques. The ANN model was rigorously trained and validated employing the stratified k-fold cross-validation method, demonstrating exceptional proficiency in detecting starch-adulterated milk samples and effectively differentiating among various starch concentrations. The principal findings indicate that the model achieved 100% accuracy, coupled with high levels of precision, sensitivity, and F1-scores. Future research should explore the application of this model to different types of adulteration and extend its validation on larger datasets. Furthermore, the potential adaptability of this method for other food products and field applications warrants investigation. This study offers valuable insights for milk producers, food safety professionals, and consumers, particularly highlighting the implications for small-scale rural farms, thereby enriching the discourse on food safety within short food supply chains.

The tactile modality is an important source of human experience and emotional expression, either on its own or by intensifying and complementing other senses, influencing our interactions with objects, people, animals and other beings. Following this, developed haptic devices transmit information to the user using tactile stimuli to increase or change sensory input. Haptics are an important factor that makes virtual worlds and remote interpersonal interaction tangible. Haptic feedback consists of more components that make an experience physically perceptible and realistic. Haptic feedbacks are widely used in mobile and wearable devices to convey various types of notifications to users. In this study, it was aimed to develop a new generation of wearable gloves against the hypoesthesia problem by combining artificial intelligence and thermohaptic, which are popular in many fields.