Computational Methods And Machine Learning
Biomedical systems are complex and often include many inputs and outputs that are difficult or impossible to measure. Computational methods offer an alternative approach to study biomedical systems by representing the system's behavior using mathematical and computational models. These models allow us to understand the system's response to different inputs without performing potentially dangerous experiments in humans or animals. Moreover, computational models and algorithms facilitate the study of different diseases by providing a mechanistic approach to understanding the disease's effect on the system's output and the impact of various treatments or procedures.
Researchers at Florida Tech use computational models to understand the human circulatory system, predict the effect of treatment on heart function, and study how neurological diseases such as stroke affect movement.
Experimental/computational mechanics of soft materials (Artery, Brain, Optic Never Head, Polymers, Composites, etc) with focus on the interface mechanics.
- Multiscale / multiphysics modeling
- Material / microstructure characterizations of soft tissue/cell
- Structure-function relationships within non-disease and diseased tissues
- Tissue/Cell remodeling
- Image-based tissue quantification
- Biomaterial failure initiation and growth
- Blast-structure interaction, traumatic brain injury
- Design/optimization of medical devices and the treatment strategies
The movement estimation and analysis laboratory focuses on understanding how neurological diseases and trauma affect movement. Our vision is to develop new technology to facilitate the diagnosis, monitoring, and treatment of neurological disorders such as stroke and Parkinson's disease.
We use computational models and machine learning to estimate movement and analyze large datasets to identify differences between healthy controls and individuals with diseases. This information is used to create predictive models that facilitate disease detection and estimation of severity.
We are also investigating different approaches for remote assessment of neurological diseases that facilitate self-assessment and reduce potentially fragile individuals' exposure to the dangers associated with clinical settings.
In our lab, we focus on understanding blood flow (hemodynamics) under healthy and pathological conditions to understand, detect, diagnose and treat cardiovascular disease. We work closely with clinicians, thereby ensuring that our research is always patient-focused. We use a combination of tools such as computational fluid dynamics (CFD) modeling of blood flow, virtual surgery and optimization, reduced order modeling of the cardiovascular system, predictive modeling, 3D printing and rapid prototyping. I envision a future where personalized medicine will deliver on its promise and I plan to contribute towards it by incorporating synergistic and interdisciplinary technologies that improve patient outcomes.
Dr. Kaya's research interests focus on cardiovascular research, ultrasound imaging and therapeutics, biosensors and medical devices. He has worked on projects including Human Patient Simulator, a seizure and epilepsy detector, a medical device to stop excessive menstrual bleeding, ultrasound image-guided release of oxygen to hypoxic tumor to enhance cancer radiotherapy, stem cells and ultrasound to treat vascular injury due to stent placement and non-invasive detection of cardiac arrhythmia to identify which patients are at the highest risk of sudden death. Specific areas of interest include developing innovative techniques and devices for the detection and therapy of cardiovascular diseases such as myocardial ischemia, cardiac arrhythmia, hypertension, hemorrhagic shock and procedures including angioplasty/stent placement and hemodynamic monitoring. In addition, Dr. Kaya has also interest in the development of contrast agents and ultrasound technology for diagnosis and therapy of diseases including cancer.