Eye Tracking during reading provides information about how readers process English texts when English is their second language (L2). We use unsupervised learning to derive reader profiles from eye tracking and comprehension data in MECO L2 dataset. For each participant, we compute eight features: skipping, regression in, refixation, reading rate, progressive duration, rereading duration, lookback duration, and comprehension accuracy. After standardization of these calculated features, a k-means clustering is deployed with and without reading rate to separate the influence of speed from other aspects of reading behavior. A gradient-boosted tree model serves as a surrogate classifier to interpret the resulting three profiles, which differ in speed, regression rate, and accuracy. We discuss how this profiling pipeline assists in studying individual differences in L2 reading and relates cluster-level patterns to existing reading style typologies.

This study integrates Bundesen’s Theory of Visual Attention (TVA) with task-evoked pupillary responses (TEPR) to examine how typog- raphy affects letter recognition. Twenty-one participants completed a whole-report task with Danish letters in three fonts (Cambria, Garamond, Roboto)× two styles (Regular, Italic) across eight expo- sure durations (10–200 ms). Bayesian posterior analysis revealed that italic processing-speed penalties scale with the degree of structural redesign: Garamond’s script-derived italic imposed the largest cost (Δ= 27.0 elem/s, (Δ > 0)= 1.00), Cambria’s glyph substitutions a moderate cost (Δ= 16.8, = 1.00), and Roboto’s oblique the smallest (Δ= 4.5, = 0.82, CI spanning zero). Pupillary metrics showed no significant font effects; only exposure duration drove dilation, though dPPD revealed a significant Font× Style interaction. This dissociation font-sensitive TVA parameters but font-insensitive pupillary am- plitude suggests typography modulates encoding efficiency rather than cognitive effort, although pupillometric

While Generative AI models like Large Language Models (LLMs) are capable of generating extensive text, their efficacy in producing readable content for human participants in experimental settings remains to be evaluated. Further, eye-tracking technology is increas- ingly utilized to study cognition and behavior, yet its application to readers’ cognitive processes when exposed to AI-generated versus human-authored texts remains unexplored. This study investigates how text generated by LLMs influences reading by analyzing gaze patterns. The study collects gaze data from 13 participants as they read AI- generated and human-authored passages. A comparative analysis is conducted within subjects to assess gaze patterns between authors and between text types based on the robust two-means clustering (I2MC) algorithm to identify fixations. In addition, pupil dilation and reading speed were examined. Our findings reveal significant differences in fixation character- istics not only between authors but also between AI-generated and human-authored texts.

While adaptive reading interfaces are capable of providing flexible typographical adjustments in real-time, readers are challenged to keep track of the context. This paper aims to contribute by introducing context preservation, which enables readers to resume reading faster after applying typographical adjustments, using eye-tracking. Typography adjustments are applied through so-called interventions, and the reading application currently has four intervention designs: Popup, Undo, Notification, and Gradual. To explore how much text is required to resume reading quickly, context-preservation functionality was applied and evaluated on 22 participants through within-subjects experiment design. Our findings reveal significant differences in reading-resume time (RRT) between interventions. Furthermore, context-preservation in a gradual intervention mode is the fastest and most liked intervention design by the participants.

Colon capsule endoscopy (CCE) as a novel 2D biomedical image modality based on visible light provides a higher perspective of the potential gastrointestinal lesions like polyps within the small and large intestines than the conventional colonoscopy. As the quality of images acquired via CCE imagery is low, so the artificial intelligence methods are proposed to help detect and localize polyps within an acceptable level of efficiency and performance. In this paper, a new deep neural network architecture known as AID-U-Net is proposed. AID-U-Net consists of two distinct types of paths: a) Two main contracting/expansive paths, and b) Two sub-contracting/expansive paths. The playing role of the main paths is to localize polyps as the target objectives in high resolution and multi-scale manner, while the two sub paths are responsible for preserving and conveying the information of low resolution and low-scale target objects. Furthermore, the proposed network architecture provides simplicity so that the model can be deployed for real time processing. AID-U-Net with an implementation of a VGG19 backbone shows better performance to detect polyps in CCE images in comparison with the other state-of-the-art U-Net models like conventional U-Net, U-Net++, and U-Net3+ with different pre-trained backbones like ImageNet, VGG19, ResNeXt50, Resnet50, InceptionV3 and InceptionResNetV2.

Background: Direct detection of polyps from colon capsule endoscopy (CCE) videos is an ultimate goal not only for physicians but also for biomedical engineers who are working on automatic internal lesions like polyps. There is also a great enthusiasm among biomedical professionals to make advanced systems for aiding doctors to have a faster and accurate diagnosis by the means of polyp detection from CCE acquired video streams. Such systems must be able to localize polyps correctly and extract the whole lesions from the video frames completely.Material and Methods: In this paper, a new approach toward object-wise polyp detection from CCE frames in a video stream is proposed. The proposed method employs modified region proposal CNNs to localize the existing polyps from CCE acquired video frames and after that a level-set method known as Distance Regularized Level Set Evolution (DRLSE) is employed for automatic model-based segmentation of localized polyps. The pixel-wise detection of polyps is necessary for polyp classification and will help gastroenterologists to determine appropriate prognosis and treatment for the patients.Results and conclusion: The proposed method is trained by the means of an CCE still image dataset which includes manually annotated polyps. The trained network is then applied to CCE video images. The results demonstrate that the proposed method is able to localize and detect polyps both region-wise and pixel-wise with a good rate of accuracy.

We introduce a novel coarse ridge orientation smoothing algorithm based on orthogonal polynomials, which can be used to estimate the orientation field (OF) for fingerprint areas of no ridge information. This method does not need any base information of singular points (SPs). The algorithm uses a consecutive application of filtering- and model-based orientation smoothing methods. A Gaussian filter has been employed for the former. The latter conditionally employs one of the orthogonal polynomials such as Legendre and Chebyshev type I or II, based on the results obtained at the filtering-based stage. To evaluate our proposed method, a variety of exclusive fingerprint classification and minutiae-based matching experiments have been conducted on the fingerprint images of FVC2000 DB2, FVC2004 DB3 and DB4 databases. Results showed that our proposed method has achieved higher SP detection, classification, and verification performance as compared to competing methods.