This study presents a specialized system for detecting textual data in medical images, aimed at enhancing the anonymization of sensitive patient information. Conventional OCR tools, such as EasyOCR and docTR, exhibit notable limitations in medical settings - particularly in handling complex imagery such as lung CT scans - resulting in high false positive rates and suboptimal recall. To overcome these challenges, we adapted the YOLOv11n object detection model for the task of text localization in medical images. The model was trained on a hybrid dataset comprising real CT images with annotated labels and synthetically generated samples, improving robustness under various imaging conditions. In a comparative evaluation, YOLOv11n achieved a precision of 100.00%, recall of 92.09% and F1 score of 95.86%, significantly outperforming EasyOCR (F1 = 66.9%) and docTR (F1 = 89.58%). In particular, YOLOv11n did not produce false positives in the test set, which makes it highly effective in preserving relevant diagnostic information while accurately detecting sensitive text regions. These results underscore the potential of adapting state-of-the-art object detection frameworks such as YOLO to specialized tasks in medical image processing and privacy preservation.

Nail disorders such as onychomycosis and nail psoriasis are common conditions that can significantly impact a patient's quality of life. Early detection and proper classification are essential for timely treatment; however, access to dermatological care may be limited. To address this challenge, we propose a mobile application designed to provide users with a preliminary diagnosis of nail plate disorders based on photographic input. The Android application uses deep learning to classify nail images. The core of the diagnostic system is trained on the publicly available “Nail Disease Image Classification Dataset” from Kaggle, which includes annotated images spanning three primary classes: healthy nails, nail psoriasis, and onychomycosis (fungal nail infection). Several CNN architectures were evaluated, with MobileNetV3Small selected for deployment due to its efficiency, achieving 93.0% test accuracy. The chosen model demonstrated high robustness and consistent performance across all three categories, making it suitable for real-world application. Moreover, MobileNetV3Small maintained a low validation loss of 0.1544 and a test loss of 0.2100, further confirming its generalization capabilities. This AI-powered mobile solution offers a user-friendly interface for self-assessment and could serve as a valuable tool for preliminary dermatological screening. Additionally, it has the potential to support healthcare professionals by assisting in triaging cases and raising awareness of potential nail disorders among patients.

The aim of this study was to develop a system capable of recognizing textual data in medical images and identifying which labels contain private information. We designed an intuitive, modular pipeline with extensibility in mind, allowing for the seamless integration or replacement of individual components as technologies evolve. This architecture ensures long-term adaptability to future advancements in OCR and machine learning. For text detection on images, we used a YOLO model fine-tuned on a specialized dataset of lung CT images. To improve text analysis, we integrated a large language model (LLM) into the workflow. Although effective, LLM introduces noticeable latency when processing numerous detected text regions, which may impact the overall image processing time. Nevertheless, with an accuracy of approximately 80% achieved on the test set, the system demonstrates robustness and scalability for the detection and interpretation of textual data in medical imaging contexts.

The objective of this study is to develop and assess a mobile application that leverages artificial intelligence (AI) to support the rehabilitation of individuals with facial nerve paralysis. The application features two primary functionalities: assessing the paralysis severity and facilitating the monitoring of rehabilitation exercises. The AI algorithm employed for this purpose was Google's ML Kit “face-detection”. The classification of facial nerve palsy was achieved by measuring the asymmetry of the user's face using a proprietary algorithm developed specifically for this study. This approach not only enables a precise assessment of paralysis severity but also allows for a personalized rehabilitation experience. Furthermore, the monitoring of rehabilitation exercise adherence and correctness is conducted through algorithms crafted for this application to ensure that patients are performing their prescribed rehabilitation exercises effectively. This comprehensive system offers a tailored and interactive approach to the management of facial nerve paralysis through the integration of AI algorithms and user-friendly mobile technology.