4 48 4. End-to-end optimization of prosthetic vision (Canny, 1986) and a deep learning-based contour detection method, called holistically nested edge detection (Xie & Tu, 2017). Differences in reconstruction performance were tested for significance using a paired t-test on the minibatches of the validation dataset. The significance level αwas set to 0.05 and adjusted with a Bonferroni correction to correct for multiple comparisons. The results of experiment 3 are displayed inFigures 4.6and4.7andTable4.3. Bothfor the perceptual reconstruction tasks and the supervised semantic boundary reconstruction task, the model seems to have adopted a different phosphene encoding strategy compared to the intensity-based reconstruction task. The average MSE is significantly lower in the intensity-based reconstruction task, and the average FSIM is significantly higher in the perceptual reconstruction task. In the supervised semantic boundary condition, 69.7% of the boundary pixels were classified correctly. Compared to the reconstructions of the SPV-encodings using existing approaches, our end-to-end model achieved adequate reconstruction performance: our model scored the highest performance for the intensity-based reconstruction (MSE, SSIM, FSIM) and the perceptual reconstruction (SSIM and FSIM). In the semantic boundary reconstruction task, our model scored average for the accuracy, sensitivity, specificity and precision, and had the largest area under the receiver-operator curve (AUC) compared to the other models. Figure 4.5: Comparison between different values of d for the perceptual reconstruction task that was used in experiment 3, whered indicates the layer depth for the VGG-based feature loss. 4.3.5. Experiment 4 Similar to earlier work in the field of simulated prosthetic vision (Bollen et al., 2019a; Bollen et al., 2019b; Dagnelie et al., 2007; Parikh et al., 2013; Sanchez-Garcia et al., 2020; Vergnieux et al., 2014; 2017), the previous experiments in this study are performed using a basic simulation of cortical prosthetic vision, with homogeneously-distributed, equallysized phosphenes. In reality, however, the exact phosphene coverage that is achieved in prosthetic vision will depend on many factors, including the electrode placement and the cortical anatomy of the patient, which is variable across people. Both early human trials (Brindley & Lewin, 1968), as well as recent animal studies (Schiller et al., 2011) have shown that individual phosphenes elicited by stimulation in V1 have different sizes, which increase with foveal eccentricity. Srivastava et al. (2009) developed a more biologically
RkJQdWJsaXNoZXIy MTk4NDMw