Advanced search
Start date
Betweenand

Development of Mobile App for calculation of biometric parameters for ophthalmic lens fitting

Grant number: 19/27836-8
Support Opportunities:Research Grants - Innovative Research in Small Business - PIPE
Duration: September 01, 2021 - August 31, 2024
Field of knowledge:Physical Sciences and Mathematics - Computer Science - Computer Systems
Principal Investigator:Marcos Makoto Ikegame
Grantee:Marcos Makoto Ikegame
Host Company:Optarium Comércio de Artigos de Ótica Eireli - ME
CNAE: Comércio varejista de artigos de óptica
Desenvolvimento e licenciamento de programas de computador não-customizáveis
City: Campinas
Associated researchers: RENATO TADAYOSHI HIRAKAWA
Associated research grant:18/01189-3 - Landmark detection and anthropomorphic measurements for calculation of biometric parameters used in ophthalmic lens fitting, AP.PIPE
Associated scholarship(s):23/12942-2 - Creation, Manipulation and Analysis of 2D and 3D Database (Data Science), BP.TT
22/16199-0 - Development of Mobile App for calculation of biometric parameters for ophthalmic lens fitting, BP.TT
22/09993-1 - Development of Mobile App for calculation of biometric parameters for ophthalmic lens fitting, BP.TT
+ associated scholarships 22/07954-9 - Development of Mobile App for calculation of biometric parameters for ophthalmic lens fitting, BP.TT
22/07214-5 - Development of Mobile App for calculation of biometric parameters for ophthalmic lens fitting, BP.TT
22/04550-4 - Development of Mobile App for calculation of biometric parameters for ophthalmic lens fitting, BP.TT
22/01457-3 - Development of Mobile App for calculation of biometric parameters for ophthalmic lens fitting, BP.TT
21/13663-4 - Development of mobile app for calculation of biometric parameters for ophthalmic lens fitting, BP.TT
21/12224-7 - Development of mobile app for calculation of biometric parameters for ophthalmic lens fitting, BP.TT
21/10231-6 - Development of mobile app for calculation of biometric parameters for ophthalmic lens fitting, BP.TT
21/09931-3 - Development of Mobile App for calculation of biometric parameters for ophthalmic lens fitting, BP.TT - associated scholarships

Abstract

According to the Brazilian Board of Ophthalmology [Ottaiano et al. 2019], 30% of the world population under 40 either need or will need prescription glasses due to the increasing prevalence of nearsightedness and astigmatism. In Brazil, where population has reached 208 million, numbers are daunting, and the portion of the population who suffer from myopia stand between 23 and 74 million; those who suffer from farsightedness stand at around 71 million (34% of the population); and those afflicted with presbyopia (also known as aging eyes condition) are 100% of the population above 55 years old (18,3% of the population). Therefore, we can posit that properly measuring lens fitting distances is of utmost importance. However, most of the population still have their measurements taken manually, through techniques that employ tools patented in the beginning of the XX century. Alternative computer-based solutions exist, but face limitations to its widespread adoption: they either rely on fixed references or make use of dedicated hardware and need a human operator who manually chancels users' facial landmarks. Extending positive results achieved on PIPE's Phase I, in which, with a 95% success rate, we proved our hypothesis that is possible to calculate, without any fixed references, the main biometric parameters needed for accurate lens fitting, the goal of Phase II is to: (i) calculate, with precision, single vision, varifocal and astigmatism lens fitting measurements without the need for fixed guidelines or external references and using technology and hardware easily available; (ii) calculate additional measurements that improve adaptation to ophthalmic lenses and improve visual acuity: Pantoscopic Tilt and Frame Wrap Angle. Such measurements demand 3D modelling of one's facial structure. Both solutions will be delivered bundled in a single product: the Lenscope APP, supported by a robust, scalable and safe client-server infrastructure. In order to refine the model we have been developing, we will adopt MTL (Multi-task Learning Models); mode precisely, our approach will be built upon Deep Learning techniques and Convolutional Networks. Regarding 3D reconstruction, we will combine MTL models to different specialist networks: (i) firstly we intend to employ a monocular approach, where the departure point is a 2D image from which a 3D model is generated; such model has recently gained ground thanks to advances in hardware and machine learning techniques, which allow for 3D structures and characteristics to be learned and derived. (ii) secondly, we plan on capturing multiple facial images from different angles, where detection needs features (landmarks) and dense feature correspondence along different positions to assure a high degree of precision. And in this case, the higher the number of corresponding features, the better features are tracked. As a result, it is expected that our product appeal to a considerable market as, in Brazil alone, 125 million need vision correction and account for a market that grosses R$ 22 billion every year. By offering a safe, scalable and user-friendly solution which runs on most smartphones in the country, we aim to achieve R$ 120 million in revenue in 5 years. (AU)

Articles published in Agência FAPESP Newsletter about the research grant:
More itemsLess items
Articles published in other media outlets ( ):
More itemsLess items
VEICULO: TITULO (DATA)
VEICULO: TITULO (DATA)

Please report errors in scientific publications list using this form.
X

Report errors in this page


Error details: