This thesis has been focused on improving access to eyecare through the design and development of portable medical devices that use low cost components, and the implementation of advanced algorithms that allows the detection and diagnosis of different eye diseases.
Uncorrected refractive errors (UREs) are a global issue impacting over 900 million individuals. This issue is particularly pronounced in low resource settings, where access to vision care is limited due to a shortage of eyecare professionals (ECP) capable of providing accurate eyeglass prescriptions. Although autorefraction-based prescriptions are a practical approach to addressing the global problem of UREs, there is always a measurable difference between the autorefractor measurement the clinical gold standard, the subjective refraction (SR). To try to reduce the variability between objective and SR measurements we have assessed the performance of machine learning (ML) ensemble models for predicting patient SR using demographic factors, wavefront aberrometry data, and measurement quality related metrics taken with the QuickSee aberrometer. The ML model significantly outperformed the autorefractor, reducing in a ±0.63 D, ±0.14 D, and ±0.08 D the 95% limits of agreement of the error distribution for M, J0, and J45, respectively. These results suggest that machine learning effectively enhances accuracy when predicting a subjective refraction of a patient from objective measurements acquired with a low-cost portable device.
A novel AI-based approach for the detection of manifest keratoconus from wavefront aberrometry images has been evaluated. The model consists of two components: a first convolutional neural network (CNN) trained to classify individual wavefront maps as either keratoconic positive or keratoconic negative, and a second component that leverages the dynamic nature of the QuickSee device. This second component provides the final classification label for a measurement based on the most frequent prediction among the individual predictions contained in the video. The algorithm demonstrated remarkable performance, achieving a 100% accuracy rate across training, validation, and test subgroups. This represents an advance in keratoconus detection based solely on total wavefront aberrometry data.
We have designed a new portable wavefront autorefractor, the QuickSee Free, that has been evaluated using an alternative study protocol that considers the variability of the SR. The proposed protocol compares the differences between the device and the SR performed by two independent ECPs to the differences between the SR measurements performed by the two ECPs, thus providing a measure of the device’s accuracy against the inherent variability of the SR procedure. The average difference between the device and each of the two SRs were found to be smaller than those found between the two SRs for the M, and comparable for the astigmatic components. Overall, the device had a high level of agreement with both SRs but was found to be more comparable to SR2 than to SR1. These results indicate the potential of the device to be used as a reliable tool for clinical and screening purposes.
In conclusion, this research aims to address the global problem of UREs and presents innovative solutions. These solutions include the design of the QuickSee Free device and various ML algorithms, all of which work towards improving access to eyecare, increasing the accuracy of eyeglass prescriptions, and detecting keratoconus.

El queratocono es una enfermedad ocular degenerativa que causa distorsión visual al afectar la córnea. La detección temprana resulta crucial, sin embargo, los métodos actuales son costosos e inaccesibles. En este trabajo, se propone un sistema de detección de queratocono accesible y fácil de usar basado en el análisis de imágenes mediante técnicas de aprendizaje automático.
Para capturar las imágenes, se utilizó el dispositivo QuickSee (PlenOptika Inc, MA, USA), mientras que para el análisis se emplearon redes  neuronales convolucionales. Además, se aplicó un proceso de limpieza de datos utilizando filtros diseñados con técnicas de procesamiento de imágenes para eliminar las imágenes no deseadas, lo cual resultó en la eliminación del 12% de los datos originales. En cuanto a las arquitecturas de las redes neuronales convolucionales, se adoptaron dos enfoques
diferentes: uno utilizando modelos pre-entrenados (Transfer Learning) y otro diseñado desde cero. Ambos enfoques obtuvieron una alta tasa de exactitud, superando el 95% en todos los casos, y una sensibilidad del 92.35% en el caso del modelo sin Transfer Learning, al considerar imágenes de forma individual. Además, al tomar todas las imágenes de cada paciente y quedándose con la predicción mayoritaria, se logró una tasa de acierto del 100% en las pruebas de diagnóstico de los pacientes del conjunto de test.
Estos resultados demuestran que este sistema de detección precoz del queratocono es una herramienta muy prometedora que podría implementarse en el dispositivo QuickSee en un futuro trabajo,
mejorando así la detección global de queratocono y su impacto en la salud ocular a nivel mundial.

At a global level, there is a critical need for technological tools that improve access to vision care to address the most common cause of visual impairment: uncorrected refractive errors. Due to the worldwide shortage of vision care professionals,obtaining accurate prescriptions for glasses that effectively correct refractive errors and restore vision is often a challenge, especially in resource-limited settings.
This thesis describes the development of advanced ophthalmological technology capable of prescribing glasses accurately, affordably, and quickly. To achieve this, we have utilized a wavefront aberrometry-based autorefractor called QuickSee, which is portable and can provide eyeglass prescriptions in just 10 seconds. Its validation in different populations and global usage has served as the starting point for this work to further improve access to vision care worldwide. Since there is always a percentage of patients considered outliers when comparing the objective results of autorefractors to the gold standard, which is the subjective refraction process where a doctor tests and asks the patient about their comfort with different lenses, we have implemented a series of advanced features to enhance these predictions. A new algorithm has been developed that accurately simulates the patient’s vision with different corrections and facilitates the determination, through iterative calculations, of the correction that maximizes certain visual quality metrics (e.g., sharpness or contrast). Additionally, a new device, QuickSee Free Pro Keratometry, has been designed, which is portable and very low-cost, improving features such as ease of use, alignment, reduced patient cooperation in measurements, and reduced weight. This device will allow individuals with minimal training to easily obtain accurate and reliable glasses prescriptions anywhere. It will integrate a high-precision wavefront aberrometer for measuring refractive errors and a keratometer for measuring corneal curvature into a compact chassis. Throughout this doctoral thesis, a new version of measurement technology has been incorporated to rapidly and accurately obtain objective refraction, along with new software developed to classify different corneal curvature radii, providing the user with more detailed information about the eye and corneal astigmatism. This new device, QuickSee Free Pro Keratometry, has the potential to address the massive problem of access to corrected vision by simplifying and expediting the glasses prescription process, potentially revolutionizing the way refraction is performed globally.