Machine Learning Approaches in Metabolic Pathway Predictions and Drug-Target Interactions: Advancing Drug Discovery

Published: Jan. 1, 2025

Language: Английский

---

Classification of OCT Images of the Human Eye Using Mobile Devices

Applied Sciences, Journal Year: 2025, Volume and Issue: 15(6), P. 2937 - 2937

Published: March 8, 2025

The aim of this study was to develop a mobile application for Android devices dedicated the classification pathological changes in human eye optical coherence tomography (OCT) B-scans. process is conducted using convolutional neural networks (CNNs). Six models were trained during study: simple network with three layers, VGG16, InceptionV3, Xception, Joint Attention Network + MobileNetV2 and OpticNet-71. All these converted TensorFlow Lite format implement them into application. For purpose, best parameters chosen, taking accuracy, precision, recall, F1-score confusion matrix consideration. designed OCT images developed Kotlin programming language within Studio integrated development environment. With application, can be performed on an image chosen from user’s files or acquired photo-taking function. results are displayed networks, along respective times each associated undergoing task. has been tested various smartphones. testing phase included evaluation score considering factors such as acquisition method, i.e., camera gallery.

Language: Английский

---

Revolutionizing Periodontal Care: The Role of Artificial Intelligence in Diagnosis, Treatment, and Prognosis

Applied Sciences, Journal Year: 2025, Volume and Issue: 15(6), P. 3295 - 3295

Published: March 18, 2025

This review evaluates the application of artificial intelligence (AI), particularly neural networks, in diagnosing and staging periodontal diseases through radiographic analysis. Using a systematic 22 studies published between 2017 2024, it examines various AI models, including convolutional networks (CNNs), hybrid generative adversarial (GANs), transformer networks. The analyzed diverse datasets from panoramic, periapical, imaging techniques, assessing diagnostic accuracy, sensitivity, specificity, interpretability. CNN models like Deetal-Perio YOLOv5 achieved high accuracy detecting alveolar bone loss (ABL), with F1 scores up to 0.894. Hybrid demonstrate strength handling complex cases, such as molars vertical loss. Despite these advancements, challenges persist, reduced performance severe limited for loss, need 3D integration. AI-driven tools offer transformative potential periodontology by rivaling clinician performance, improving consistency, streamlining workflows. Addressing current limitations large, advanced techniques will further optimize their clinical utility. stands poised revolutionize care, enabling early diagnosis, personalized treatment planning, better patient outcomes.

Language: Английский

---

Bifunctional bowl-shaped organosilica for efficient exosome capture and metabolites panel based hepatocellular carcinoma diagnosis

Chemical Engineering Journal, Journal Year: 2025, Volume and Issue: unknown, P. 163980 - 163980

Published: May 1, 2025

Language: Английский

---