MEDIA LAB / MITCertain types of blindness might be avoided if inexpensive, informative diagnoses were made available in less wealthy areas. This is the case of cataracts. A white cloud that spreads over the eye and that keeps light out, slowly jeopardizing vision. Despite the progress of surgical techniques, the disease continues to be one of the main causes of blindness in the world. However, technology developed at the Massachusetts Institute of Technology (MIT), in the United States, with fundamental participation by Brazilians, may help to get this disease detected easily. Connected with the process of aging, cataracts are the cause of blindness in 18 million people worldwide, according to the World Health Organization (WHO). The innovation is a device with a system to detect cataracts that can be attached to a mobile phone. The work won two major awards in May: MIT Ideas, a contest that has an expert panel of judges and that focuses on social projects that involve needy communities, and the MIT Global Challenge Public Choice Award, in which people who are not within the academic community vote on projects. The project is to be presented next August at the ACM Siggraph computing conference. The device is both compact and inexpensive and is based on what the person being diagnosed can see. “It can be used to screen patients in places where there is limited access to ophthalmology,” said Manuel Oliveira, a professor at the Information Technology Institute of the Federal University in Rio Grande do Sul (UFRGS), one of the Brazilians involved in the project and who was an associate visiting professor at MIT from 2009 to 2010. Bringing together knowledge of graphic computing, optics and interactivity techniques, along with a good understanding of mathematics, the system, named Catra, allows one to detect the presence, site and seriousness of cataracts, generating maps that devices currently available in consulting rooms cannot produce. “We believe that Catra will give rise to a revolution in high precision, affordable devices, in which the patients have the right to see, file and understand the raw data of their own health, monitoring and testing their vision anywhere,” said Ramesh Raskar, whose brainchild this project was, and leader of the Camera Culture group of the Media Lab of MIT, where the work was conducted.
Dot of light
“The basis of the technology is to use the mobile phone’s screen as a source of light and an interactive software program, developed by the researchers, that sweeps over different points of the eye in search of alterations, blockages or dispersions of light,” says Oliveira. Every instant the light appears in a different part of the screen. To direct the rays of light projected by the mobile phone, they created a plastic device with a small hole and a lens. When the dot of light appears on the mobile phone screen, the light goes through the orifice and turns into a ray, a light filament. The lens within the device ensures that the rays reach the cornea and the crystalline lens, the natural lenses of the eye; then they converge to a single point on the user’s retina. Every time the dot of light moves on the mobile phone screen, the ray of light enters through a different part of the cornea.
The user looks at the mobile phone screen through the device and responds to different commands by pressing the keys on the device itself. People who have no cataract will see, in the first test, which is characterized by an automatic sweeping of the eye, a green dot whose intensity remains unchanged and is steady and does not twinkle. If the person does have the disease, however, the instant the ray of light passes through the white mark of the cataract, the dot vanishes quickly. This happens because the cataract causes the ray of light to change its course, both blocking and spreading the light.
In the second stage, an interactive software program beeps for each point tested in the natural lenses of the patient’s eye. Should the ray disappear, or become dark, one of the mobile phone buttons has to be pressed, marking the region as one affected by cataract. This is repeated several times to assess the consistency of the responses, generating, at the end, a map on which the points where the person noticed some luminosity change, indicating the location and size of the cataract. Now that the software knows where the cataract points are, the next stage is to choose one of them and compare it to a region where the lenses are clear and free of deformations. Using the same device, the patient compares the two luminous paths and using the mobile phone buttons, makes one point look like another one, generating a small map. With this, one can tell to what extent the mark is blocking out light. This is equivalent to the density of the cataract.
The Catra project began when Oliveira and his doctoral student Vitor Pamplona arrived at the Media Lab in 2009 wanting to combine their studies of mathematical models geared toward the physiology of the human eye with Raskar’s knowledge of camera technology. The early trials used a network of microlenses arranged along a piece of paper, with colored dots beneath the lenses, for the researchers to understand where the light from each lens got to. These were the earliest stages of what would become the Netra (Near-Eye Tool for Refractive Assessment), an apparatus for prescribing glasses similar to Catra, that can detect myopia, hypermetropia and astigmatism via a device attached to a mobile phone. Pamplona took part in the two studies; in the second, with the collaboration of Erick Passos, from Federal Fluminense University. “The invention works, but its clinical use is still questioned by physicians, because we didn’t do a broader clinical validation, comparing results with detailed medical information of the users,” says Pamplona.
“The work has some merit and is interesting, but it’s still too early to apply it clinically,” says the ophthalmologist Rubens Belfort Filho, from the Federal University of São Paulo (Unifesp). Belfort would like to see researchers in the medical area taking part in the scientific articles published by the group and is concerned that the test is subjective. “There isn’t yet a scientific and objective means of evaluating the progress of cataracts. That is bad, because if I’m going to conduct clinical research, I want to know whether a drug is good enough to stop the progress of cataract or not. However, this method is also not good, because it’s based on what the patient thinks he is seeing,” he says.
Shrikant Bharadwaj, associate director of optometry of the L V Prasad Eye Institute (LVPEI), an eye health treatment and research network in India and one of the WHO centers for preventing blindness, says that a device such as the Catra would have a huge impact in his country. “At present, the dimension of the cataract is assessed using a lantern, which does not detect all types of the disease, only the most advanced ones. The Catra would be much more quantitative,” he explains. Structured with centers of excellence and care centers, LVPEI serves the population in rural areas in the so-called primary vision centers that cover, on average, a population of 20 thousand village inhabitants. Those who are identified as having the disease may be referred to a more complex treatment center if necessary. For Bharadwaj, a device like the Catra could be used by healthcare professionals on a door-to-door basis, to measure cataracts in the population. The other option would be to use a slit lamp, a common piece of equipment in medical consulting rooms, used for ophthalmological exams (it is the one on which we rest our chin while the physician assesses our eyes). The problem is that both the fixed and the portable versions of the device are very expensive.
Bharadwaj says that the issue of patient interactivity with the mobile phone is a challenge that they are still dealing with, especially in rural populations that use the Netra to prescribe glasses. “The results tend to include more noise and the time it takes to train patients who live in villages to be able to do the test using a mobile phone is not ideal yet,” he says. That is why Bharadwaj traveled to MIT in May, to discuss with the group led by Rasker ways of making the test less subjective. The LVPEI team does not yet know whether the same difficulty found in using the Netra in the field will also apply to the Catra. They have just received a prototype of the latter. “Now we’re organizing ourselves to test the Catra on individuals from a rural environment,” says Bharadwaj.
Meeting of multidisciplinary ideas, creativity and innovation
The laboratories at the MIT Media Lab resemble architecture firms, where engineers, artists, scientists and designers work in such different fields as neuroengineering, cars of the future, and musical compositions. Based on a lot of creativity, prototypes are built and tested in open spaces, on benches where the collaborative spirit and “hands on” approach of MIT can be observed. “Taking to pieces and re-assembling is an important activity at the Media Lab,” says Manuel Oliveira, a professor at the Information Technology Institute of the Federal University of Rio Grande do Sul (UFRGS) who was a visiting professor at the Media Lab from 2009 to 2010. Glass partitions separate the researchers from curious pedestrians who walk by the facility, which has become a tourist attraction in Boston. The laboratory was the birthplace of such companies as E-ink, which launched the first generation of screens for e-books (Kindle); Harmonix, which markets the Rock Band and Guitar Hero games; and the non-profit organization One Laptop Per Child (OLPC), co-founded by Nicholas Negroponte, a professor at MIT. Negroponte founded the Media Lab in 1985, together with Jerome Wiesner, then dean of MIT. Negroponte’s argument was that architectural methods should be used to develop new technologies and new media. Anyone who clicks on the interactive screens located in the Media Lab building can become acquainted with 30 research groups working on more than 400 projects. The groups are comprised of 28 professors and researchers, 139 graduate students, and visiting professors. The source of funding is a pool of companies that pay to “try out projects that are too far-fetched or risky for them,” says Leo Burd, a Brazilian researcher currently working at the Media Lab. Oliveira points out that people tend to over-romanticize the lab. “Not everything that is being done there is spectacular, but the facility attracts very competent people interested in building up an environment of experimentation and hard work.”
Last April, Japan’s Joi Ito, who lives in the United States, was appointed as the new director of Media Lab. Ito, who had never worked in the academic environment, helped to establish Japan’s first internet service – in 1994 – and is the chairman of the board of Creative Commons, an organization whose objective is to share digital information. Ito, who also invests in technology companies, helped found the Twitter, Flickr and Last.fm. sites.Republish