A5030177073- Cutaneous Melanoma Detection and Management
- 3D Shape Modeling and Analysis
- Medical Imaging and Analysis
- Computer Graphics and Visualization Techniques
- Quality and Safety in Healthcare
- Medical Image Segmentation Techniques
- Optical Coherence Tomography Applications
- Advanced X-ray and CT Imaging
- Virtual Reality Applications and Impacts
- Advanced Numerical Analysis Techniques
- Augmented Reality Applications
- Radiomics and Machine Learning in Medical Imaging
- Optical Imaging and Spectroscopy Techniques
- Management of metastatic bone disease
- Digital Image Processing Techniques
- Experimental Learning in Engineering
- AI in cancer detection
- Medical Imaging Techniques and Applications
- Robotics and Automated Systems
- Nutritional Studies and Diet
- Infrared Thermography in Medicine
- Manufacturing Process and Optimization
- Anatomy and Medical Technology
- Artificial Intelligence in Healthcare and Education
- Thermoregulation and physiological responses
Riga Technical University
2010-2024
The modern education environment is changing rapidly, as the technology that being utilized for educational purposes. Emerging advanced technologies such artificial intelligence (AI), machine learning, cloud computing, and virtual reality (VR) are reshaping systems creating new ways to digitalize modernize a generation of learners. In this paper, structural design an adaptable system proposed distance learning support higher courses, aiming increase degree interactivity informational...
The integration of artificial intelligence (AI), particularly through machine learning (ML) and deep (DL) algorithms, marks a transformative progression in medical imaging diagnostics. This technical note elucidates novel methodology for semantic segmentation the vertebral column CT scans, exemplified by dataset 250 patients from Riga East Clinical University Hospital. Our approach centers on accurate identification labeling individual vertebrae, ranging C1 to sacrum–coccyx complex. Patient...
The choice of technique for the creation a 3D digital human bone model from natural specimens has critical impact on final result and usability obtained model. cornerstone factor in modeling is number faces polygon mesh, along with topological accuracy, as well resolution level detail texture map. Three different techniques (3D scanning, photogrammetry, micro-computed tomography) have been used to create zygomatic bone. As implementation use models can be divided into three main...
The incidence of skin cancer is still increasing mostly in industrialized countries with light- skinned people. Late tumour detection the main reason high mortality associated cancer. accessibility early diagnostics Latvia limited by several factors, such as cost dermatology services, long queues on state funded oncologist examinations, well inaccessibility oncologists countryside regions - this an actual clinical problem. new strategies and guidelines for post-surgical follow-up intend to...
According to the EU statistics1 skin cancer – melanoma is one of most deadly cancers. At same time, it has up 95% chance2 for curing if detected on early stage. The diagnostics obstructed by low availability dermatology specialists and percentage patients that are taking regular diagnostics. To deal with unavailability we propose implementing portable automated diagnostic device, available wide range medical institutions. Automated systems not yet capable providing final diagnosis, but...
The research presents image quality analysis and enhancement proposals in biophotonic area. sources of problems are reviewed analyzed. with most impact area analyzed terms specific task – skin cancer diagnostics. results point out that main problem for is the illumination problems. Since it often not possible to prevent problems, paper proposes post processing algorithm low frequency filtering. Practical show diagnostic improvement after using proposed filter. Along that, filter do reduces...
The objective of this study was to develop and evaluate artificial intelligence (AI) models for the detection instance segmentation vertebrae spinal metastases in computer tomography (CT) scans. were trained on datasets consisting patients with polytrauma relatively undamaged spines, as well diagnosed metastases. Our results indicate that achieved high performance vertebra segmentation, F-beta scores ranging from 0.88 0.96 across all classes. For metastases, model attained 0.68 lytic type...
Objectives: The integration of machine learning and radiomics in medical imaging has significantly advanced diagnostic prognostic capabilities healthcare. This study focuses on developing validating an artificial intelligence (AI) model using U-Net architectures for the accurate detection segmentation spinal metastases from computed tomography (CT) images, addressing both osteolytic osteoblastic lesions. Methods: Our methodology employs multiple variations architecture utilizes two distinct...
The goal of our study is to train artificial neural networks (ANN) using multispectral images melanoma. Since the number melanomas limited, we offer synthesize them from benign skin lesions. We used previously created melanoma diagnostic criterion p'. This calculated lesions captured under 526nm, 663nm, and 964nm LED illumination. these three nevus so that p' map matches criteria (the values in lesion area >1, respectively). Demonstrated results show by transforming possible get a reliable...
The paper proposes an approach of a novel non-contact optical technique for early evaluation microbial activity. Noncontact will exploit laser speckle contrast imaging in combination with artificial neural network (ANN) based image processing. Microbial activity process comprise acquisition time variable patterns given sample, ANN processing and visualization obtained results. proposed technology measure (like growth speed) implement these results counting live microbes. It is expected, that...