Skin diseases are one of the most common health problems that affect people of all ages around the world and significantly reduce the quality of life of individuals. Diseases of eczema, seborrheic dermatitis and skin cancer need to be diagnosed and correctly classified promptly. This issue, which is of great importance in terms of control and practical and effective treatment, is the study's starting point. The study included 693 individuals with eczema, 750 with skin cancer and 770 with seborrheic dermatitis. In the study, which focused on the classification of 3 different skin diseases, the Relief algorithm was used to increase the classification success and to ensure the selection of more meaningful qualities. With AlexNet with cross-validation, the accuracy rate was 89.39% for 80% training and 20% test rates. When SVM classification with the Relief algorithm was used for the same rates, the accuracy rate was 92.10%. In the analysis performed on the ISIC 2017 dataset, the accuracy rate is 89.16% for 80% training and 20% test rate. When the training and test rate was changed to 70% training and 30% test rate, the accuracy rate was 91.11%. It was observed that SVM classification with Relief's algorithm offers higher accuracy rates than other methods. The proposed model provides an original contribution to the literature, particularly through its integration of feature selection and a simplified architecture. This high success rate reveals that deep learning is an effective method in classifying skin diseases and the transfer learning process and will reduce the mortality rates due to cancer diseases with early and effective treatment while enabling skin diseases to be easily distinguished.

Monitoring the freshness of meat and fish is crucial for enhancing food safety for consumers and protecting public health. In this study, a paper sensor using bromothymol blue indicator sensitive to colour change was developed. The sensor performance was compared in solutions prepared with different dye amounts and different holding times. Studies were conducted on meat and fish, and colour changes were analysedanalyzed using both RGB and ΔE values. In the first sensors produced on fish that had not yet come into contact with food, RGB values resembled grey-greenish tones. In spoiled samples, the colour changes to reddish tones. RGB analysis of meat transitions from beige to reddish tones in intact samples and from beige in spoiled samples. RGB is a method that provides objective results in monitoring spoilage. The findings emphasize that colour changes are not merely visual differences but can also be used as a highly effective parameter in determining product condition. Furthermore, solutions containing higher BTB showed more pronounced and rapid colour changes, particularly in response to alkaline compounds formed during spoilage. The paper sensor developed for determining the freshness of meat and fish products offers a practical, low-cost, rapid, and intuitive alternative. The study contributes to the optimization of sensor systems using pH-sensitive synthetic dyes and presents RGB, L*, a*, and b*-based data.

Asymptomatic in its early stages. This situation increases the need for automated and reliable screening systems, especially those based on retinal fundus images. This study presents a hybrid classification framework for glaucoma detection that combines classical image processing methods with deep learning-based representations, possessing the potential to support clinical processes and exhibiting generalizability. In the proposed approach, deep features obtained from the fc7 layer of a pre-trained AlexNet architecture, along with HSV colour histograms, Haralick textural features, LBP and HOG descriptors, are combined using the ORIGA retinal fundus dataset. The class imbalance problem was addressed by applying the SMOTE method exclusively to the training data, while the high-dimensional feature space was optimised using the MultiSURF algorithm. The selected features obtained were evaluated using Support Vector Machines (SVM), Random Forest (RF), and XGBoost classifiers. Experimental results demonstrate that the proposed hybrid framework can accurately distinguish between individuals with glaucoma and those without, with high accuracy. Specifically, the Random Forest model demonstrated the highest performance with 96.92% accuracy, an F1 score of 0.939, and an AUC value of 0.994. The findings reveal that the combined use of deep and hand-crafted features, supported by MultiSURF-based feature selection and balanced learning strategies, offers a robust and stable solution for glaucoma detection. The proposed approach is considered an automated glaucoma screening method with the potential to be integrated into clinical decision support systems for early diagnosis.

Automatic detection of smoking behavior is of great importance to public health. High classification accuracy is reported in deep learning, but how this works and how it relates to various biases are difficult to explain. To address this gap, a smoking detection model based on Segment Anything Model (SAM), incorporating evolutionary neural networks and Gradient Weighted Class Activation Mapping (Grad-CAM) analyses to suppress background noise, has been proposed. In a study of 1120 images, an accuracy of 88.84% and a Receiver Operating Characteristic - Area Under the Curve (ROC-AUC) value of 0.9613 were obtained. Furthermore, data augmentation strategies were included to improve the robustness and generalization performance of the model, and statistical significance analysis was added to show that the performance improvements provided by the SAM-based approach were not due to random variation. At the same time, the false negative rate, a crucial concept for the health field, was relatively reduced. The analyses confirmed that there is an agreement between the model and hand-to-mouth-smoking interactions. It offers a focused, interpretable, reliable, and practical solution for enclosed spaces.