Multi-view deep learning for rigid gas permeable lens base curve fitting based on Pentacam images
- PDF / 1,532,844 Bytes
- 16 Pages / 595.276 x 790.866 pts Page_size
- 60 Downloads / 186 Views
ORIGINAL ARTICLE
Multi-view deep learning for rigid gas permeable lens base curve fitting based on Pentacam images Sara Hashemi 1 & Hadi Veisi 1 & Ebrahim Jafarzadehpur 2 & Rouhollah Rahmani 3 & Zainabolhoda Heshmati 1 Received: 22 June 2019 / Accepted: 5 March 2020 # International Federation for Medical and Biological Engineering 2020
Abstract Many studies in the rigid gas permeable (RGP) lens fitting field have focused on providing the best fit for patients with irregular astigmatism, a challenging issue. Despite the ease and accuracy of fitting in the current fitting methods, no studies have provided a high-pace solution with the final best fit to assist experts. This work presents a deep learning solution for identifying features in Pentacam four refractive maps and RGP base curve identification. An authentic dataset of 247 samples of Pentacam four refractive maps was gathered, providing a multi-view image of the corneal structure. Scratch-based convolutional neural network (CNN) architectures and well-known CNN architectures such as AlexNet, GoogLeNet, and ResNet have been used to extract features and transfer learning. Features are aggregated through a fusion technique. Based on a comparison of means square error (MSE) of normalized labels, the multi-view scratch-based CNN provided R-squared of 0.849, 0.846, 0.835, and 0.834 followed by GoogLeNet, comparable with current methods. Transfer learning outperforms various scratch-based CNN models, through which proper specifications some scratch-based models were able to increase coefficient of determinations. CNNs on multi-view Pentacam images have enabled fast detection of the RGP lens base curve, higher patient satisfaction, and reduced chair time. Keywords Transfer learning . Multi-view convolutional neural network . Image analysis . RGP lens base curve fitting
1 Introduction Rigid gas permeable (RGP) contact lenses have been widely used for patients prone to keratoconus and irregular astigmatism [1, 2]. The use of these lenses is the primary method of vision correction in the mentioned conditions [3]. However, * Hadi Veisi [email protected] Sara Hashemi [email protected] Ebrahim Jafarzadehpur [email protected] Rouhollah Rahmani [email protected] Zainabolhoda Heshmati [email protected] 1
Faculty of New Sciences and Technologies, University of Tehran, Tehran, Iran
2
Department of Optometry, School of Rehabilitation Science, Iran University of Medical Sciences, Tehran, Iran
3
Digikala Company, Tehran, Iran
the lens fitting process requires long ophthalmologist and patient chair time in addition to its difficulty. This is due to various existing techniques, distinctive lens designs, and difficulty in matching the surface of the cornea and RGP lens [4]. Ophthalmologists start the fitting process based on the information provided in the patient’s Pentacam four refractive maps [5]. The Pentacam uses non-invasive rotating Scheimpflug imaging technology for anterior and posterior measurements [6]. The corneal topography influences the evalu
Data Loading...
