Carla Agurto

At present time, I am working on algorithms to extract informative features from speech based on phonation, articulation, among others, to detect candidate profiles potentially linked to medical conditions. I am currently applying these techniques to Alzheimer's disease, Parkinson's disease (PD) and placebo-controlled studies. In addition, I am working in the characterization of dyskinesia, tremor, and other PD symptoms using sensor data.

My previous experience was mainly focused in medical imaging analysis. I worked on several projects related to the detection of retinal diseases such as: diabetic retinopathy (DR), glaucoma, papilledema, age-related macular degeneration (AMD), malarial retinopathy, cardiovascular diseases, retinopathy of prematurity. Since the aims of these projects were to be implemented in a screening system, I also worked on the development of image quality software and the improvement of low quality images using new enhancement techniques. My latest work in this area was to combine medical laboratory data with fundus images to increase detection rates of different eye diseases (DR, AMD, and glaucoma) resulting in an awarded grant. I also worked in the application of level set methods for the segmentation of ultrasound cardiac images and the characterization and detection of diabetic peripheral neuropathy using infrared videos of the feet after applying a cold stimulus. I also have experience in the implementation of algorithms in field- programmable gate arrays (FPGAs.) which was the main focus of my thesis 'A Fixed-Point Implementation of the CORDIC Algorithm in FPGAs'