Projects / Programmes
Biomedical Image and Signal Analysis
January 1, 2022
- December 31, 2027
| Code |
Science |
Field |
Subfield |
| 2.06.00 |
Engineering sciences and technologies |
Systems and cybernetics |
|
| 1.02.00 |
Natural sciences and mathematics |
Physics |
|
| Code |
Science |
Field |
| 2.06 |
Engineering and Technology |
Medical engineering
|
| 1.03 |
Natural Sciences |
Physical sciences |
Biomedicine, images, signals, automated analysis, digital holographic microscopy, quantitative evaluation, membrane, vesicle, cell, tissue, organ, biomarkers, diagnosis, image-guided procedures, therapy pathology artificial intelligence, deep learning
Data for the last 5 years (citations for the last 10 years) on
April 23, 2024;
A3 for period
2018-2022
| Database |
Linked records |
Citations |
Pure citations |
Average pure citations |
| WoS |
737 |
18,905 |
15,314 |
20.78 |
| Scopus |
897 |
23,743 |
19,315 |
21.53 |
Researchers (17)
Organisations (1)
Abstract
Biomedical image and signal analysis have become one of the most important tools for studying and interpreting different phenomena in biology and medicine. Over the past decades, many revolutionary techniques and devices for acquiring, storing, analyzing and displaying biomedical images and signals have been introduced, which allow research scientists and medical practitioners to obtain qualitative insights as well as quantitative evaluations of their data, which in turn support their scientific hypotheses, and aid in clinical observations and medical diagnoses. Nevertheless, biomedical image analysis, which is focused on different imaging modalities like microscopic techniques, multi- and hyperspectral imaging, X-ray imaging, computed tomography, magnetic resonance, ultrasound, etc., as well as biomedical signal analysis, which is focused on different signal modalities like electrocardiograms, electroencephalograms, electromyograms, etc., still continue to expand at a fast rate. More than ever, previously unknown information about structure and function of normal and disease-affected organelles, cells, tissues, organs, organ systems and organisms is being discovered. Results and (clinical) adaptation strategies related to the computational and quantitative analysis of biomedical images and signals have so far lagged behind the corresponding acquisition techniques. Nevertheless, there has been a substantial increase in the resulting performance quality, primarily due to increase in the number and size of databases (e.g. big data), and the application of novel artificial intelligence techniques (e.g. deep learning) to identify, classify and quantify features and patterns in biomedical images and signals. Consequently, we are witnessing a remarkable increase of interest and efforts to apply the results in practice. In medicine, automated image and signal analysis is being applied at all stages of patient treatment, from studying normal and pathological states, diagnosis, treatment planning and simulation, and image-guided therapy, to monitoring of the disease progression and evaluation of treatment outcomes. Research will be primarily focused on: A) medical image analysis and B) digital holographic microscopic (DHM) imaging. In the field of biomedical image analysis, we will design, develop and validate deep learning supported image registration, segmentation and quantitative evaluation techniques with clinical applications on X-ray, CT and MR images. These techniques will aid in the interpretation of image information and provide support to image-assisted medical examinations, treatment decisions and follow-up evaluations. Research in the field of DHM imaging will be focused on the development of novel optical systems for snapshot tomographic imaging and flow cells for high-throughput analysis of individual particles and cells, and development of related deep learning-aided light propagation, phase retrieval and reconstruction methods.
Significance for science
Within the proposed research field of biomedical image analysis, we will design, develop and evaluate computer-assisted algorithms based on state-of-the-art technologies such as artificial intelligence, specifically the cutting edge concepts of deep learning. These technologies will allow to enhance common biomedical image analysis tasks such as image registration and segmentation, but also explore exciting new possibilities such as knowledge distillation and discovery in the domain of image-based medical diagnosis and prognosis. In particular, relevant contributions to engineering and medical sciences will be one of the important results of the proposed research. New findings are expected to result in advances in the field of computerized medical image analysis as well as in the fields of biomedical engineering, medical imaging and biomechanics. One of the several impacts on science will also be the contribution to the fundamental understanding of cell motion and identification of critical cytoskeleton and cell membrane mechanical properties/descriptors that define the ability of a cell to move. All these may set future directions of research on cell movement in biology and medicine. An important objective of the proposed project is to create a universally applicable theoretical framework for controlled interactions of viruses, drugs and nanoparticles with biomembranes, based on the prediction of distribution and the number of topological defects on the cell membrane surface with specific geometries/topologies. Deep understanding of such interactions will aid the design of biocompatible and hybrid liposome-based drugs. Knowledge on cell shape transformations and membrane nanovesiculation may have an important impact on the future development of novel therapeutic methods using nanovesicles as biogenic nanocarriers for genetic material, peptides, synthetic drugs etc. for the treatment of diseases such as inflammations, cancer and Covid-19. Our research results will be presented in the form of publications, presentations, data, prototypes, educational material, policy recommendations, knowledge and skills and will paw the way for novel research, collaboration with other research groups and companies, new or improved products and services, and improved education, all with the goal to bring benefit to science, economy, citizens/society and environment.
Significance for the country
Impact on economy: The proposed research outcomes in the field of biomedical image and signal analysis fall into the category of medical technology, as they will represent the application of knowledge and skills in the form of computer-assisted techniques that will be developed to aid in health care, and could improve the quality of life and the general well-being. These findings will be reflected in the application of novel ideas and methods, and potential transfer and implementation of knowledge into national and foreign healthcare institutions and business communities. We will put every effort into exploiting the results and thus develop and create new or improved products and services that will be marketed by companies that we have already founded (Sensum, Inteliteh), companies that we might found in future, or other established high-tech companies. In the past years, the industry has shown an increased interest in image-based applications, and therefore the research outcomes may also be attractive for software developers and medical device manufacturers. Impact on citizens and society (health): Medical image examinations are nowadays an essential part of the clinical workflow, and therefore the acquisition and analysis of medical images is indispensable considering the increasing number of images being daily acquired worldwide. In the long run, the results of the proposed research could enable artificial intelligence to improve diagnostic accuracy and reduce the diagnosis-to-treatment turnaround time, and consequently relieve the heavy burden of medical professionals when evaluating or treating different diseases. The other important impact on health will be the contribution to the development of novel therapeutic strategies that can target specific organs and cells, which is one of the main objectives of modern medicine. Impact on environment: Poor air quality, associated with high levels of particulate matter (PM), is one of the five leading health risks, alongside high blood pressure, smoking, diabetes and obesity, and it poses a risk that is inherently difficult to avoid. Black carbon PM emissions are also largely responsible for the positive radiative forcing by aerosols, being the second most important human emission in terms of its climate forcing after CO2. The novel digital holographic microscopy-based methods that we will developed for quantification of PM in real-time could be exploited to improve the modeling and understanding of the effects that aerosols have on the global radiation balance through the absorption and scattering of light. One of the key aspects of the proposed research is also to develop novel technologies and methods for next generation compact low-cost quantitative measurement systems. Such systems could be deployed in large numbers, creating a dense worldwide grid of PM sensors. The rich quantitative stream of information provided by such a grid could be used to improve source apportionment that is essential for effective management of pollution sources through informed policy decision-making, better understanding of the adverse health effects of PM pollution, better understanding of the climate processes and empowering the scientific community with a new powerful set of quantitative tools.
