Fatma Akalın

Imaging techniques are widely used in dentistry to understand the 3D structure of teeth and detect diseases, but their interpretation is time-consuming and prone to error. To address this, decision support systems are increasingly utilized. This study proposes a CNN-based classification model using the UFPE dataset, which includes Cone Beam Computed Tomography (CBCT) scans. In the first scenario, both real and enhanced images were input into a CNN, yielding 68.92% accuracy for enhanced images. Due to a result, enhanced images were used in all other scenarios. In the second scenario, a newly designed CNN architecture called ToothNet, incorporating a custom activation function, was tested. It achieved 69.92% accuracy, 61.45% recall, 62.67% precision, and 68.68% F1-score, showing a 1.45% increase in accuracy. To evaluate generalizability, three more classification scenarios were examined using the same dataset. ToothNet achieved 80.14% accuracy in the “healthy vs. large lesion” and 68.73% in the “healthy vs. small lesion” classification. These results indicate that the proposed architecture not only improves accuracy but is also generalizable across different lesion sizes.

Artificial intelligence is a framework that enables human cognitive abilities and intelligent behaviors to be performed by machines. It has strong potential to make successful inferences and offer quick solutions for diseases in the field of health. Integrating this potential into the detection and diagnosis process of monkeypox, which is characterized as a pandemic case, is valuable in terms of controllability of the epidemic. Because monkeypox is a disease with a high transmission rate. In addition, the clinical features of the disease are easily confused with skin diseases such as measles, chickenpox and eczema. On the other hand, since symptoms usually appear about 3 weeks after the virus enters the body, contact with respiratory droplets, bodily fluids or infected objects, which play an important role in the spread of the disease during this period, will increase the rate of spread. The declaration through the Centers for Disease Control and Prevention that 99,518 cases have been confirmed as of August 2024 confirms this situation. In this regard, artificial intelligence solutions stand out for providing fast and accurate diagnosis. For this, various deep learning models (VGG16, MobileNetV2, ResNet50, DenseNet169, InceptionV3, XceptionV3, EfficientB7, AlexNet) were used to classify the Mpox Skin Lesion Dataset Version 2.0 dataset in the first stage. In the second stage, six different regularization methods were proposed to optimize performance. In the third stage, the success rate was optimized by integrating the improved hybrid loss function. In the fourth stage, the improved approach was applied to the augmented dataset. The results show that the proposed approach improves the accuracy rate by approximately 6% compared to the original classification architecture.

Objectives: In recent years, artificial intelligence (AI) has become a reliable technology in clinical decision support systems with the solutions it offers in the field of medicine. This finding highlights the effectiveness of monkeypox (Mpox) detection. Early diagnosis, timely intervention, and controlled disease management are the support that the informatics world can offer for Mpox. Methods: A review study was conducted to analyze AI-assisted studies on Mpox for identifying disease clusters, tracking cases, predicting future outbreaks, determining mortality risk, diagnosing and managing diseases, and studying disease trends. The studies reviewed in this review for the detection of Mpox disease are generally shaped around machine learning algorithms, deep learning-based methods, hybrid techniques, and ensemble learning approaches. Results: The advantages and disadvantages of each method introduced in this review are detailed, as well as the choice that should be made for the target goal depending on the parameters of hardware capacity, size of the dataset, training time, computational cost, purpose of the model, and application area of the model. Conclusions: This review provides a comprehensive study for researchers and practitioners with solutions to current and future challenges within the framework of the pros and cons of AI technology for Mpox disease.

Acute lymphoblastic leukemia (ALL) is a critical type of cancer affecting white blood cell development. It is characterized by the disruption of normal stem cell maturation processes or the excessive proliferation of leukemic cells. As the most common childhood cancer, ALL is a significant health problem requiring continuous clinical monitoring. A major challenge in diagnosing the disease stems from the non-specific nature of the initial symptoms. Common symptoms, such as fever, fatigue, headache, weight loss, and musculoskeletal disorders, can mimic many diseases, making diagnostic accuracy challenging. This ambiguity leads to delayed diagnosis and negatively impacts treatment success. Although the incidence of leukemia has increased in recent years, advances in medical technology have shown promising results in reducing mortality rates. Health information systems play a significant role in this success by facilitating early disease diagnosis. In our research, two different image enhancement methods were developed to highlight the characteristics of blast cells, a critical indicator for the diagnosis of ALL. Three image groups were evaluated: unprocessed original images and two different processed image sets (versions 1 and 2). Classification performed using the MobileNetv2 transfer learning framework achieved accuracy rates of 85%, 90%, and 89% on the test dataset, respectively. To enhance diagnostic reliability beyond single-model performance, an ensemble architecture combining eight different transfer learning networks was created. This optimized ensemble model, using images enhanced with the high-performing version 1 filter, achieved approximately 90% classification accuracy. This result represents a significant advancement in AI-assisted hematological image interpretation.

Due to the increasing complexity of industrial systems, fault detection hinders the continuity of productivity. Also, many methods in industrial systems whose complexity increases over time have a mechanism based on human intervention. Therefore, the development of intelligent systems in fault detection is critical.. Avoiding false alarms in detecting real faults is one of the goals of these systems. Modern technology has the potential to improve strategies for detecting faults related to machine components. In this study, a hybrid approach was applied on two different datasets for fault detection. First, in this hybrid approach, data is given as input to the artificial neural network. Then, predictions are obtained as a result of training using the ANN mechanism with the feed forward process. In the next step, the error value calculated between the actual values and the estimated values is transmitted to the feedback layers. IWO (Invasive Weed Optimization) optimization algorithm is used to calculate the weight values in this hybrid structure. However the IWO optimization algorithm is designed to be initialized with fractal-based weighting. By this process sequence, it is planned to increase the global search power without getting stuck in local minima. Additionally, fractal-based initialization is an important part of the optimization process as it keeps the overall success and stability within a certain framework. Finally, a testing process is carried out on two separate datasets supplied by the Kaggle platform to prove the model’s success in failure detection. Test results exceed 98%. This success indicates that it is a successful model with high generalization ability.

Artificial intelligence has made revolutionary advances in medical imaging in recent years. Various algorithms and techniques are used in this scientific field to significantly improve the accuracy and speed of medical diagnosis and classification processes. In this direction, approaches have been improved, from the past to the present, to extract meaningful features from dental images and classify them accurately. Especially, high asymmetry in morphological balance, play a critical role in distinguishing pathological patterns from normal anatomy. In this study, we propose a scenario for the classification of periapical lesions, supported by a combination of improved image processing techniques and regularization strategies integrated into the VGG16 transfer learning architecture, as the experience and time criteria required for manual interpretation of lesion detection confirm the need for a computer-aided system. In this study, which was conducted on the UFPE public dataset, an improvement in the performance of the VGG16 transfer learning architecture was achieved, with 18 different regularization methods proposed. These values indicate optimized training within the parameters of avoiding overfitting, stability, generalizability, and high accuracy. This optimization has the potential to use as a decision support system for diagnosis and treatment processes in various subfields of the medical world.

The eye is one of our five sense organs, where optical and neural structures are integrated. It works in synchrony with the brain, enabling the formation of meaningful images. However, lack of function, complete absence or structural abnormalities of cone cells in the cone cells in the retina causes the emergence of types of Color Vision Deficiency (CVD). This deficiency is characterized by the lack of clear vision in the use of colors in the same region of the spectrum, and greatly affects the quality of life of the patient. Therefore, it is important to develop filters that enable colors to be combined successfully. In this study, an original filter design was improved, built on a five-stage systematic structure that complements and supports itself. But optimization regarding performance value needs to be tested with objective methods independent of human decision. Therefore, in order to provide performance analyses based on objective evaluation criteria, original and enhanced images simulated by patients with seven different Color Vision Deficiency (CVD) types were classified with the MobileNet transfer learning model. The classification results show that the developed final filter greatly improves the differences in color perception levels in both eyes. Thus, color stimulation between the two eyes is more balanced, and perceptual symmetry is created. With perceptual symmetry, environmental colors are perceived more consistently and distinguishably, and the visual difficulties encountered by color blind individuals in daily life are reduced.

Electrocardiography (ECG) is a non-invasive tool used to identify abnormalities in heart rhythm. It is used to evaluate dysfunctions in the electrical system of the heart. It offers a mechanism that does not cause any harm to patients. Being affordable makes it accessible. It provides a comprehensive assessment of the condition of the heart. Although it provides a successful analysis opportunity for arrhythmia detection, it is time-consuming and depends on the clinician’s experience. In addition, since the ECG patterns in pediatric patients are different from the ECG patterns in adults, physicians consider it a difficult and complex task. For this reason, a custom dataset of pediatric patients was created in this study. This dataset consists of 1318 abnormal beats and 1403 normal beats. MobileNetv2 transfer learning architecture was used to classify this balanced dataset. However, the stability of the results is a valuable. Therefore, the optimization algorithm that minimizes the loss function and the regularization method that controls the complexity of the model are proposed. In this direction, Proposed Optimization Algorithm V5 and Proposed Regularization Method V5 approaches have been integrated into the MobileNetv2 transfer learning model. The accuracy rates produced in the training and test datasets are 0.9801 and 0.9509, respectively. These results have acceptable improvement and stability compared to the accuracies of 0.9633 and 0.9399 produced by the original MobileNetv2 architecture on the training and test dataset, respectively. However, performance values provide limited information about the generalizability of the model. Therefore, the same processes were repeated on a more complex dataset with 6 categories. As a result of the classification, the accuracy rates for the training and test data sets were obtained as 0.9200% and 0.8975%, respectively. Training was performed under the same conditions as the training performed on 2-category datasets. Therefore, it is normal for the test dataset to experience a decrease of approximately 5%. The results obtained show that generalizations can be made for comprehensive, highly diverse and rich datasets.

Machine learning is a powerful decision support system used in analyzing and evaluating real-life data. This system aims to create new solutions and improve performance. Therefore, it is related to the field of data science. There are data on the basis of this relationship The effectiveness of drawing meaningful insights from data depends on the quality of the model's training. To improve this performance, the variety of combinations among the data and the total number of data in the dataset should be increased. But in this topic, insufficient data access, legal regulations, ethical rules, confidentiality procedures, privacy, data sharing restrictions and cost parameters are obstacles. Synthetic data generation is a basic step in the field of data science in order to solve all these problems, improve functionality and provide powerful machine-learning inferences. Therefore, a new synthetic data generation approach consisting of 3 basic stages is proposed in this study. In the first stage, synthetic data production similar to the distribution of the original data was carried out with the modified ABC (Artificial Bee Colony) optimization algorithm. In the second stage, the category information of the independent variables was determined by the statistical evaluation analyzed with regression methods among the artificial data produced. In the third stage, the efficiency and applicability of the artificial data produced were evaluated with supervised machine learning classifiers. As a result of the evaluation, it has been proven that the proposed synthetic data generation approach improves the performance of machine learning classifiers in proportion to the increasing number of data. The decision tree algorithm that showed maximum performance produced success rates of 100%, 92.5%, 100%, 85%, and 66% on 5 separate enriched datasets, respectively.

Pamukkale University Journal of Engineering Sciences