The aim of this systematic review is to compare the efficacy of convolutional neural networks (CNN) and Vision Transformers (ViT) in the field of dental imaging, in order to examine in depth the potential, advantages, and limitations of both models in this domain. The search strings used in the study were “((“Vision Transformer” OR ViT OR “Transformer architecture”) AND (“Convolutional Neural Network” OR CNN OR ConvNet) AND (Dental OR Dentistry OR “Maxillofacial” OR “Oral Radiology”) AND (Image OR Imaging OR Radiograph))”. The search was conducted in January 2025. Two investigators independently evaluated the full texts of all eligible articles and excluded those that did not meet the inclusion/exclusion criteria. Of 2596 articles, 21 met the inclusion criteria. Depending on the task category, of the 21 studies that were reviewed, 14 (66.7%) utilized classification, while 7 (33.3%) utilized segmentation. Panoramic radiography is the most commonly used imaging modality (52.3%) and the ViT-based model was observed to have the highest performance (58%). ViT-based deep learning models tend to exhibit higher performance in many dental image analysis scenarios compared to traditional convolutional neural networks. However, in practice CNN and ViT approaches can be used in a complementary manner.

Climate change and its impact on agricultural production due to the occurrence of extreme weather events appear to be more imminent and severe than ever, presenting a global challenge that necessitates collective efforts to mitigate its effects.There have been many practical and modelling studies so far to estimate the extent of climate change and possible damages on agricultural production, suggesting that water availability may decrease by 50% and agricultural productivity between 10 and 30% in the coming years ahead. Though there have been many studies to estimate the possible level of damage by the climate change on the production of many agricultural crops, no study has been conducted on the greenhouse tomato production. Therefore, this study was conducted to discover the effects of extreme high temperatures during the 2022–2023 growing season on the high-tech Turkish tomato greenhouse industry through a survey. The results showed that all greenhouses lost yield, ranging from 6 to 53%, with an average of 12.5%. Survey data revealed that irrigation and fog system water consumption increased by 29.32% and 31.42%, respectively, while fertilizer and electricity consumption rose by 23.66% and 19%. Some 76.5% of the growers declared difficulty in climate control, 11.7% reported tomato cluster losses with no information on yield loss, 9% experienced yield losses despite no cluster losses, and 61.7% observed a decline in tomato quality, leading to reduced sales prices. Considering these findings, it is recommended that greenhouses must adopt advanced climate control technologies, expand fog system capacities, and integrate renewable energy sources to enhance resilience against climate-induced challenges. Additionally, improving water-use efficiency, optimizing cooling strategies, using new and climate-resistant varieties and adjusting cropping seasons could help mitigate yield losses due to extreme temperatures. The study results offer extremely valuable insights into greenhouse production for researchers, technology developers, and policymakers for the mitigation of climate change effects and the development of more sustainable production systems.

Periapical radiographs (PAR) offer information about the pulp and periodontal health of teeth. However, intraoral radiographs are insufficient for diagnosing buccolingual anomalies and variations such as bifid canals due to their two-dimensional nature. Cone beam computed tomography (CBCT) is the gold standard for 3D imaging in the clinic but requires additional radiation. The aim of this study was to create a software (XPAR) which obtains x-ray depth analysis and 3D modelling of the pulps of single-rooted teeth by converting the grey values in the original radiographs into numerical data. Two single-rooted teeth were included in the experimental part of the study. Chicken fibula bone was preferred for alveolar bone simulation because it could simulate cortical and trabecular structures due to similarity. A total of four images (60kVp & 70kVp; single alveolar bone & double alveolar bone) were obtained. The aim of this experimental part is to test the repeatability and realism of the algorithm to be created for pulp modelling. Retrospectively, 31 single-rooted teeth with both periapical radiography and cone-beam computed tomography imaging were included in the retrospective part of the study. According to XPAR, depth increase areas were interpreted as root resorption and accessory canal. Depth decrease areas were evaluated as the transformation of the pulp from an elliptical to an oval form, pulp stone, bifid canal formation and the presence of thick alveolar bone. The diagnostic accuracy of XPAR application on pathological and morphological changes was evaluated by comparing the obtained results with CBCT. 80% of the analyses diagnosed as bifurcation by XPAR application were supported by CBCT. This rate decreased to 27% in the diagnosis of transitions from elliptical to oval form. A total of 5 and 19 linear formations observed in the form of depth decrease and increase, respectively, were accepted as image errors in XPAR. Buccolingual bifid canal formations and pulp obliterations can be diagnosed with a rate of nearly 50% with the depth decrease finding obtained in XPAR application. Imaging errors caused by deformed detectors are typically observed as linear formations. A software named XPAR was studied to perform x-ray depth analysis and 3D pulp modelling from radiographs. XPAR software converts radiograph grey values into numerical data for 3D modelling. Study tested two single-rooted teeth using chicken fibula for alveolar bone simulation. XPAR showed 80% accuracy for bifurcation diagnosis compared to CBCT results. XPAR diagnoses bifid canals and pulp obliterations with ∼ 50% accuracy; some image errors.

This study aimed to develop a UVC machine designed for the treatment of common fungal, bacterial, and viral diseases in greenhouse-grown crops, serving as an alternative to chemical applications. The machine utilizes standard battery carts or trolley infrastructure for mobility along heating pipes in greenhouses. The machine is equipped with 8 UVC lamps, four on each side, each having a power of 145 W and a length of 1.5 m, complete with reflectors. The lamps can be adjusted both vertically and horizontally to optimize their positioning based on the crop conditions. Two 24 V switches control the lamp power levels according to data received from the microcontroller. Each lamp can be controlled via software, allowing them to be turned on or off as needed. The lamps on either side of the machine can be operated independently to provide the specific dosage required for treating plant diseases within a range of 5 to 100 mJ/cm². UVC doses were measured at various machine speeds and lamp-sensor distances. Statistical analyses confirmed a significant effect of PWM speed on UVC dose (p < 0.05), demonstrating the inverse relationship between machine speed and accumulated UVC exposure. The UVC machine has proven effective as an autonomous device for crop disease management. Operating the lamps at slower speeds increases the dosage but may delay treatment, which is critical for stopping the spread of diseases or pests.

Objectives: This study evaluated whether panoramic radiographs could help identify C-shaped canal configurations in mandibular first premolars and offer preliminary diagnostic indicators using the MediXpar bioinformatic platform by examining pixel density and grayscale depth, in a way that approaches the diagnostic value of cone-beam computed tomography (CBCT) while minimizing radiation exposure. Methods: Panoramic radiographs were retrospectively analyzed with the MediXpar system, a pixel-based quantitative image analysis platform that extracts grayscale density and texture-related parameters from digital radiographs. Three quantitative measures—Xpar, Power Xpar, and Pixel Count—were recorded under zoomed and non-zoomed conditions at two-pixel resolutions (1023 × 496 and 1023 × 593). Logistic regression and receiver operating characteristic (ROC) analyses were used to assess the diagnostic performance of each parameter in identifying C-shaped canals C-shaped canals, with statistical significance set at p < 0.05. Results: Teeth with C-shaped canals showed a significantly larger dentin–pulp complex volume ( p = 0.001), whereas pulp length did not differ ( p = 0.653). Digital zoom produced higher pixel-based values ( p ≤ 0.01), reflecting resampling effects rather than added anatomy. Among all metrics, Power Xpar demonstrated the strongest diagnostic performance (AUC = 0.917 in zoomed and 0.950 in non-zoomed images), showing consistent discrimination across resolutions. Conclusions: Digital magnification alters the appearance of panoramic images but does not improve diagnostic accuracy. Power Xpar performed best at higher native resolution and remained highly discriminative despite on-screen zoom, indicating that its usefulness depends on true image resolution rather than digital enlargement. Clinical relevance: Although zooming enhances visual comfort, it may distort quantitative interpretation. MediXpar-based analysis, especially using Power Xpar—offers a practical and reproducible option when CBCT is unavailable.