ABSTRACT Alzheimer's disease (AD) diagnosis using MRI scans must be very accurate since the subtle differences throughout the course of the disease are difficult to identify. Traditional approaches are not effective, and new computational techniques are required that can provide fast and accurate diagnosis. In this paper, a novel deep learning methodology that greatly enhances the sensitivity and specificity of AD stage identification by analyzing in‐depth MRI scans is proposed. The model applies a novel Sequential Convolutional Neural Network (CNN) architecture, which has been deeply trained on the “Augmented Alzheimer MRI Dataset” made available by Kaggle, to integrate various layers of depth and complexity to identify and scan in‐depth features on MRI images. Major enhancements include the use of learning rate schedulers and dropout regularization to fine‐tune training as well as avoid overfitting, with a diagnosis accuracy of 94.2%. This level of accuracy not only makes diagnostic processes easier but also allows for early detection of Alzheimer's phases, which is crucial for timely interventions and effective management of the condition. The model is rigorously trained on a large set of augmented data with varying levels of AD to guarantee robustness and generalizability in various demographic and clinical settings. Batch normalization and higher‐order activation functions allow faster and stable convergence of training, and thus the model is more efficient and scalable. Application of this model to the clinic has the potential to sharply reduce time to diagnosis, lessen dependence on radiological expertise, and offer a high‐accuracy, scalable imaging device enabling early and accurate treatment in Alzheimer's care. This innovation represents a significant next phase in medical imaging with artificial intelligence, and it offers a highly effective tool for fine detection and staging of Alzheimer's disease.

ABSTRACT The surge in digitally altered images has necessitated advanced solutions for reliable image verification, impacting sectors from media to cybersecurity. This work provides an effective method of real vs. deepfake image distinction through utilization of the EfficientNetV2‐B2 model, the latest in convolutional neural networks known for its accuracy and effectiveness. The research utilized a big dataset of 100,000 images equally divided between deepfake and real classes to create a balanced sample. The methodology involved preprocessing images to a fixed size, utilizing augmentation techniques to enhance model robustness, and employing a systematic training schedule along with accuracy parameter optimization. Significantly, the research utilized an automated learning rate adjustment mechanism to optimize training performance, contributing to a complex model calibration. Outcome of the experiment design was showing 99.89% classification accuracy and an equally impressive F1 score, which is a measure of the efficiency of the model in identifying deepfakes. The results provided in‐depth analysis with some misclassifications, providing recommendations for potential image processing and model training improvements. The outcome points to the suitability of applying EfficientNetV2‐B2 where there is a requirement for high accuracy in image authentication.

The prevalence of distressing content on social media raises concerns about users' mental well-being, prompting the use of trigger warnings (TW) and content warnings (CW). However, inconsistent presentation of TW/CW across platforms and the lack of standardized practices confuse users regarding these warnings. To better understand how users experienced and utilized these warnings, we conducted a semi-structured interview study with 15 social media users. Our findings reveal challenges across three key stakeholders: viewers, who need to decide whether to engage with warning-labeled content; posters, who struggle with whether and how to apply TW/CW to the content; and platforms, whose design features shape the visibility and usability of warnings. While users generally expressed positive attitudes toward warnings, their understanding of TW/CW usage was limited. Based on these insights, we reflected on the TW/CW mechanisms from multiple stakeholders' perspectives. Lastly, we further reflected on our findings and discussed the opportunities for social media platforms to enhance users' TW/CW experiences, fostering a more trauma-informed social media environment.

With mounting privacy concerns, and movement towards a cookie-less internet, session-based recommendation (SR) models are gaining increasing popularity. The goal of SR models is to recommend top-K items to a user by utilizing information from past actions within a session. Many deep neural networks (DNN) based SR have been proposed in the literature, however, they experience performance declines in practice due to inherent biases (e.g., popularity bias) present in training data. To alleviate this, we propose an underlying neural-network (NN) based Structural Causal Model (SCM) which comprises an evolving user behavior (simulator) and recommendation model. The causal relations between the two sub-models and variables at consecutive timesteps are defined by a sequence of structural equations, whose parameters are learned using logged data. The learned SCM enables the simulation of a user's response on a counterfactual list of recommended items (slate). For this, we intervene on recommendation slates with counterfactual slates and simulate the user's response through learned SCM thereby generating counterfactual sessions to augment the training data. Through extensive empirical evaluation on simulated and real-world datasets, we show that the augmented data mitigates the impact of sparse training data and improves the performance of the SR models.

Abstract Precise monitoring and diagnosis of epilepsy by manual analysis of EEG signals are challenging due to the low doctor‐to‐patient ratio, and shortage of medical resources. To automate this diagnosis in real‐time, EEG based Brain–Computer Interface (BCI) system with integration of artificial intelligence techniques will prove to be propitious. This work proposes an end‐to‐end, one‐dimensional atrous conv‐net‐based architecture for automatic epilepsy diagnosis using EEG signals with a conceptual framework of the EEG‐BCI system for routine monitoring and clinical use. The proposed architecture has a robust backbone of six blocks of atrous convolutional layers activated with exponential linear unit functions. The six blocks are followed by the addition of a long short‐term memory layer for automatic feature extraction and sequential EEG data analysis. The efficacy of the proposed architecture has been verified on three publicly available EEG datasets using various evaluation metrics, feature maps, test set evaluation, and ablation studies. An average training and validation accuracy of 96.16% and 90.80% has been achieved upon multiple runs for the three datasets. Ablation experiments indicate that the addition of each block contributed to increasing 17%–25% accuracy scores during the classification of epileptic and non‐epileptic EEG signals. The real‐time EEG‐BCI has been analyzed using weight optimization of the proposed architecture through the NVIDIA Tensor RT framework on a 40 GB DGX A100 NVIDIA workstation. The proposed architecture has generalized well in comparison with the existing techniques for the three EEG datasets and achieved a low training and validation loss with optimum evaluation metrics. This makes the proposed architecture suitable for future EEG‐BCI system deployment in the automatic diagnosis of epilepsy.

Clinical decision making and diagnosis in healthcare is enabled by learning causal relationships among entities. In order to answer the question, ’Will changing the treatment regimen lead to quicker recovery of a patient?’ requires predicting counterfactuals using survival analysis. We seek to reliably estimate the average treatment effect in terms of survival probabilities of the factual outcomes as compared to survival probabilities of the unobserved counterfactual population. Due to the nature of observational data, this framework is prone to treatment selection bias and censoring bias. We address these issues and propose the novel SurvCI framework which consists of a counterfactual inference (CI) algorithm for data with time-to-event outcomes. We investigate the performance of the proposed framework on synthetic and semi-synthetic data and employ survival and causal metrics to empirically confirm that our method outperforms baselines.

The COVID-19 pandemic disrupted processes interaction designers took for granted, challenging some of our most commonplace design practices. Participatory and situated approaches have been impacted the most: where we engaged stakeholders in-person and in-context, during this time we must co-design remotely and in virtual environments. Such a dramatic change calls for new co-design methods. In this article, we present a novel remote strategy for involving stakeholders to co-design interactive technology: Designerly Tele-Experiences (DTE). Our methodological proposal enables participants to experience early design concepts in-the-wild as a provocation to contribute new ideas that resonate with their experiential preferences. Here we describe the rationale for DTE, unpack how it builds on and extends existing methods, and provide actionable guidelines from our experience of using it in our work. Our contribution will empower interaction designers to embrace participatory and situated approaches even when engaging stakeholders in person is not possible or desirable.

In this paper, we present an annotated portfolio of speculative ideas that emerged from a co-design process where we investigated the playful potential of day-to-day mealtime. Our portfolio illustrates the learnings from our participatory engagements: it embodies ours and our participants' ideas of how technology might support increasingly playful and socio-emotionally rich experiences around food. We contribute: (1) a list of play potentials of mealtime-i.e. people's existing playful practices with food-that will point designers towards socio-emotionally desirable play-food experiences; (2) a portfolio of speculative design ideas that illustrate how mealtime technology could help to realize that playful potential; and (3) a discussion of our participants' experiences with and responses to lo-fi prototypes of our ideas. Our work will provoke designers to carefully consider the impact of food-tech innovation on the quality of people's social eating experiences and inspire them to cultivate forms of food-play that are socio-emotionally rich.?