Dry eye syndrome may be identified more easily and inexpensively using computational techniques
In partnership with the company, ICMC researcher will develop design in video and image processing area: Goal is to create low-cost prototype to simplify the diagnosis of disease.
Sensation of sand in the eyes, discomfort, irritability, burning, redness, difficulty to stay at the computer, in air-conditioned environments and to perform routine tasks such as prolonged reading. If you are feeling these symptoms, it may be more one of about 18 million people who suffer from dry eye syndrome, the second leading cause of care in clinics according to oftlmologista and president of the Hospital Israelita Albert Einstein, Claudio Lottenberg. Making the diagnosis of this disease easier and cheaper is the challenge that is mobilizing a researcher at the Institute of Mathematics and Computer Sciences (ICMC) of USP in São Carlos.
“The idea of the project is to develop a noninvasive method. Let’s build a prototype: will be glasses with attached micro cameras and light-emitting diodes (LEDs). Through the analysis of the captured images and other resources, we can identify whether the person has the syndrome, “explains James Trojahn, doctoral student of ICMC and professor at the Federal Institute of São Paulo (IFSP), São Carlos campus. The proposal seems very simple, however, computationally, there are several obstacles to be faced so that you can make the product available on the market.
The potential of the initiative was recognized by the Support of São Paulo Research Foundation (FAPESP): the project is among the 46 selected in the 3rd Innovative Research Program cycle Small Business (PIPE) .Over the next nine months, Trojahn work with researchers from technology company and medical equipment Wavetek , based in San Carlos, to demonstrate the scientific and technical feasibility of the project and will be available to 120 thousand reais (Phase 1 of the PIPE). For this, with the support of four professionals with FAPESP grants: two designed to solve the computational challenges and two facing the development of the prototype itself (optical elements, mechanical and electronic). If the search is successful in the first stage, researchers may request more resources for the program seeking to consolidate the proposal (phase 2). FAPESP may also, at a later stage (stage 3), support the commercial and industrial development of the product.
Eyeing the eye – are the tears that lubricate the surface of our eyes, making it homogeneous and creating what experts call tear film. Dry eye syndrome occurs when there is a decreased production of tears or a deficiency in some of its components. Among the most common methods currently used to diagnose the disease are two tests. One of them consists in placing a filter paper strip into the patient’s eye ( Schirmer test ) and after five minutes, measure the amount of moisture in the strip. The other method ( Rose Bengal test ) is made by placing drops of a coloring properties and with the eye response analysis if dry spots, they will absorb the dye differently, delimiting the affected area.
Make the diagnosis much faster and comfortable for patients, and barateá it is the main goal of the study coordinated by Trojahn. The prototype will have coupled LEDs, which are the light emitting sources for the eyes as well as micro cameras to record everything that happens with patients. A computer program will be created to automatically analyze videos that will be generated, it will be necessary to evaluate all the times when there was a disturbance in the flow of light, either because there was a break in the tear film or because the person blinked.
“First, you must select the significant images that should be analyzed. When we recorded what is happening in the eye of the person, also grasp the lid of the image of the eyelashes, in addition to the disturbances that occur when a person moves. You will need to delete all the information of the images, “explains Trojahn. Select only what is relevant is essential in this work as a video contains a huge amount of data (big data ) and the more information we have, the longer it took to evaluate them.
Another aspect that makes the work of analysis is to divide the images into different parts (segmentation).”If the person winking, we can separate the images of what happened before and after, for example. So, I can see how much time has passed since the last time she blinked, “says doctoral student. According to him, the prototype also will identify where it is going to break the tear film, which makes it easier to see if there are changes in one of the lacrimal glands.
To Trojahn, the knowledge acquired in ICMC are essential for the development of the project: “I took a course on digital image processing that will be useful in the project. Although my goals in doctoral are different, the object to analyze is the same: videos. ”
How the project was born – The idea came about when the company’s founder Wavetek , Luis Alberto de Carvalho, developing his post-doctorate in the United States, the University of Rochester New York , during the years 2005 and 2006. “There was a teacher in the laboratory where I worked I was trying to simplify the diagnosis of dry eye syndrome. He had a huge and very expensive equipment which was commercially unviable and therefore never became a product available on the market, “says Carvalho. When this teacher noted that the Brazilian postdoc was sometimes with red eyes, I invite you to be one of the volunteers in your search. Result: Carvalho found that had the dry eye syndrome, but in a slight degree.
“Since that time, technology has evolved and, last year, I decided to look for partners in the imaging area to try to develop a product aimed at the diagnosis of the syndrome,” said Carvalho, who is a research associate of the Center for Research in Optics and photonics (CEPOF) , based at the Institute of Physics of São Carlos (IFSC), and also a professor of postgraduate studies at the School of Medicine / Federal University of São Paulo. He reached for Trojahn indicating one of his trainees, Numajiri Miriam, who is astudent of the Federal University of São Carlos and collaborator of the project.
Possible applications – View the equipment that will help create available in ophthalmologists’ offices and pharmacies is what motivates Carvalho and Trojahn. But doctoral student ICMC said that if the computational techniques that are searching are effective in dry eye syndrome detection may be applied to other purposes: “If it works, starts to become possible to identify other diseases such as deformation of the eyeball pressure in the eye and can even prevent eye stroke, a much more serious problem that causes blindness. ”
Another possible Search utility is conducting mass diagnostics. This is necessary, for example, in cases of chemical disasters, such as the gas leak happened in the Port of Santos last January 14 and released a toxic cloud over the affected region. “If many people have symptoms of eye irritation, we can use our equipment to identify the problem quickly,” says Trojahn.
In future, the doctoral student believes will also benefit from their own study, “How do I much time on the computer and this is one of the environmental risk factors for the development of dry eye syndrome, I’m pretty sure I’m a strong candidate to acquire the disease. ” ophthalmologist’s office out with the diagnosis without putting a tape in the eye or eye drops dripping, will be a great victory of Trojahn.
Text: Denise Casatti – ICMC Communications Department / USP
Picture Credits: Reinaldo Mizutani (photo Trojahn) and National Eye Institute