Pupil Size Prediction Techniques Based on Convolution Neural Network

Whang, Allen Jong-Woei; Chen, Yi-Yung; Tseng, Wei-Chieh; Tsai, Chih-Hsien; Chao, Yi-Ping; Yen, Chieh-Hung; Liu, Chun-Hsiu; Zhang, Xin

doi:10.3390/s21154965

Cited by 5 publications

(3 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…[3] Pupil diameter measurement has been the subject of many studies. [2,6] It is well known that pupil size decreases with increasing luminance. Factors investigated for possible relationships with pupil size are gender, iris color, age, refractive error etc.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

The influence of brightness, age and refractive errors on pupil size

Yılmaz,

Özçelik

2022

Anatomy

View full text Add to dashboard Cite

Objectives: In this study, we aimed to investigate the influence of age, gender and refractive status on pupil size under different illuminance conditions in presbyopic healthy people. Methods: A total of 75 volunteers (49 females and 26 males) were included in the study. The medical records of each participant was reviewed. Patients were divided into three groups according to their age and refractive errors. Pupil responses were evaluated with the automated pupillometry function of the Sirius Topographer. Relationship between pupil size and age, sex, laterality, dominant eye, refrative errors were analyzed by statistical analyses. Results: In the early aged presbyopia group of patients the mean value of the hyperopic group was higher than the other two groups in photopic measurements. In scotopic condition, the mean value of the myopic group was found to be higher than the emmetropic and hyperopic groups. No difference was observed in the comparison of mesopic measurements in all three groups. In the patients with established presbyopia, myopic pupil diameter was higher compared to emmetropic and hyperopic pupils in all illuminances. There was a difference only in the scotopic condition when comparing the emmetropic and hyperopic groups. Pupil size did not change with age in all three conditions in emmetropic eyes. Pupil diameter decreased in all three conditions with age in hyperopics. Photopic and mesopic pupil size increased with age in myopics. Conclusion: The results of this study support that the effects of patient age and refractive status on pupil diameter are important in optimal lens design.

show abstract

Section: Discussionmentioning

confidence: 99%

“…The vision between these two ranges is defined as mesopic vision. [2] Pupil diameter is an important element in the optical quality of the eye. Increase in pupil size increases high-order monochromatic aberrations, and this results in reduced image quality.…”

mentioning

confidence: 99%

The influence of brightness, age and refractive errors on pupil size

Yılmaz,

Özçelik

2022

Anatomy

View full text Add to dashboard Cite

show abstract

“…However, since the image will be similar to the original image, the risk of overfitting, i.e., a decrease in the performance on the test dataset due to the prediction model fitting to match into the training dataset, cannot be ruled [57][58][59][60][61][62][63][64][65][66][67][68]. Thus, data augmentation effectively enables learning with a small number of data.…”

Section: Angles and Data Split In Deepsnap-dl With Digits And Python ...mentioning

confidence: 99%

A Deep Learning-Based Quantitative Structure–Activity Relationship System Construct Prediction Model of Agonist and Antagonist with High Performance

Matsuzaka

Uesawa

2022

IJMS

View full text Add to dashboard Cite

Molecular design and evaluation for drug development and chemical safety assessment have been advanced by quantitative structure–activity relationship (QSAR) using artificial intelligence techniques, such as deep learning (DL). Previously, we have reported the high performance of prediction models molecular initiation events (MIEs) on the adverse toxicological outcome using a DL-based QSAR method, called DeepSnap-DL. This method can extract feature values from images generated on a three-dimensional (3D)-chemical structure as a novel QSAR analytical system. However, there is room for improvement of this system’s time-consumption. Therefore, in this study, we constructed an improved DeepSnap-DL system by combining the processes of generating an image from a 3D-chemical structure, DL using the image as input data, and statistical calculation of prediction-performance. Consequently, we obtained that the three prediction models of agonists or antagonists of MIEs achieved high prediction-performance by optimizing the parameters of DeepSnap, such as the angle used in the depiction of the image of a 3D-chemical structure, data-split, and hyperparameters in DL. The improved DeepSnap-DL system will be a powerful tool for computer-aided molecular design as a novel QSAR system.

show abstract

SIPFormer: Segmentation of Multiocular Biometric Traits With Transformers

Hassan

Ahmed

et al. 2023

IEEE Trans. Instrum. Meas.

View full text Add to dashboard Cite

The advancements in machine vision have opened up new avenues for implementing multimodal biometric identification systems for real-world applications. These systems can address the shortcomings of unimodal biometric systems, which are susceptible to spoofing, noise, nonuniversality, and intraclass variations. Besides, ocular traits among various biometric traits are preferably used in these recognition systems due to their great uniqueness, permanence, and performance. However, segmenting visual biometric features under unconstrained situations remains challenging due to a variety of variables, such as Purkinje reflexes, specular reflections, eye gaze, off-angle pictures, poor resolution, and numerous occlusions. To overcome these challenges, this research presents a novel framework called SIPFormer, comprising the encoder, decoder, and transformer blocks to simultaneously segment three ocular traits (sclera, iris, and pupil) using its discriminative multihead self-attention mechanism. Besides, we used the large publicly available iris database reflecting different unconstrained acquisition settings, with inherent noise effects such as scanner artifacts, intensity and illumination variations, motion blur, and occultations caused by eyelashes, eyelids, and eyeglasses. Furthermore, the simulation results demonstrate the efficacy of the proposed SIPFormer model, where it achieved the mean Dice similarity coefficient scores of 0.9018, 0.9176, and 0.9229 for segmenting the sclera, iris, and pupil classes, respectively.

show abstract

Pupil Size Prediction Techniques Based on Convolution Neural Network

Cited by 5 publications

References 37 publications

The influence of brightness, age and refractive errors on pupil size

The influence of brightness, age and refractive errors on pupil size

A Deep Learning-Based Quantitative Structure–Activity Relationship System Construct Prediction Model of Agonist and Antagonist with High Performance

SIPFormer: Segmentation of Multiocular Biometric Traits With Transformers

Contact Info

Product

Resources

About