Intelligent Assistance Research Articles

Voice Assistants for Mental Health Services: Designing Dialogues with Homebound Older Adults.

The number of older adults who are homebound with depressive symptoms is increasing. Due to their homebound status, they have limited access to trained mental healthcare support, which leaves this support often to untrained family caregivers. To increase access, a growing interest is placed on using technology-mediated solutions, such as voice-assisted intelligent personal assistants (VIPAs), to deliver mental health services to older adults. To better understand how older adults and family caregivers intend to interact with a VIPA for mental health interventions, we conducted a participatory design study during which 6 older adults and 7 caregivers designed VIPA-human dialogues for various scenarios. Using conversation style preferences as a starting point, we present aspects of human-likeness older adults and family caregivers perceived as helpful or uncanny, specifically in the context of the delivery of mental health interventions, which helps inform potential roles VIPAs can play in mental healthcare for older adults.

Virtual resonance: analyzing IPA usage intensity under COVID-19's isolating canopy

The widespread adoption of smartphones coupled with advancements in artificial intelligence has significantly propelled the use of intelligent personal assistants (IPAs). These digital assistants have become indispensable for many users, particularly during the COVID-19 pandemic. Employing coviance-based structural equation modeling (CB-SEM) and analyzing data from 260 participants, this study explores the key factors influencing IPA usage intensity. Contrary to expectations, affective risk perception showed no significant impact on either IPA usage or parasocial interaction during the pandemic. In stark contrast, cabin fever syndrome significantly influenced both IPA usage and parasocial interaction, underscoring the role of environmental and psychological stressors in shaping technology use. Furthermore, loneliness was found to significantly enhance parasocial interaction with IPAs, though it did not affect usage intensity. The findings highlight a substantial connection between parasocial interaction and IPA usage intensity, suggesting that users who engage in human-like interactions with IPAs tend to use them more extensively. These insights not only deepen our understanding of how IPAs are utilized during health crises but also point to potential directions for developing IPAs that are more responsive to users' emotional and social needs.

Interobserver Agreement and Performance of Concurrent AI Assistance for Radiographic Evaluation of Knee Osteoarthritis.

Background Due to conflicting findings in the literature, there are concerns about a lack of objectivity in grading knee osteoarthritis (KOA) on radiographs. Purpose To examine how artificial intelligence (AI) assistance affects the performance and interobserver agreement of radiologists and orthopedists of various experience levels when evaluating KOA on radiographs according to the established Kellgren-Lawrence (KL) grading system. Materials and Methods In this retrospective observer performance study, consecutive standing knee radiographs from patients with suspected KOA were collected from three participating European centers between April 2019 and May 2022. Each center recruited four readers across radiology and orthopedic surgery at in-training and board-certified experience levels. KL grading (KL-0 = no KOA, KL-4 = severe KOA) on the frontal view was assessed by readers with and without assistance from a commercial AI tool. The majority vote of three musculoskeletal radiology consultants established the reference standard. The ordinal receiver operating characteristic method was used to estimate grading performance. Light kappa was used to estimate interrater agreement, and bootstrapped t statistics were used to compare groups. Results Seventy-five studies were included from each center, totaling 225 studies (mean patient age, 55 years ± 15 [SD]; 113 female patients). The KL grades were KL-0, 24.0% (n = 54); KL-1, 28.0% (n = 63); KL-2, 21.8% (n = 49); KL-3, 18.7% (n = 42); and KL-4, 7.6% (n = 17). Eleven readers completed their readings. Three of the six junior readers showed higher KL grading performance with versus without AI assistance (area under the receiver operating characteristic curve, 0.81 ± 0.017 [SEM] vs 0.88 ± 0.011 [P < .001]; 0.76 ± 0.018 vs 0.86 ± 0.013 [P < .001]; and 0.89 ± 0.011 vs 0.91 ± 0.009 [P = .008]). Interobserver agreement for KL grading among all readers was higher with versus without AI assistance (κ = 0.77 ± 0.018 [SEM] vs 0.85 ± 0.013; P < .001). Board-certified radiologists achieved almost perfect agreement for KL grading when assisted by AI (κ = 0.90 ± 0.01), which was higher than that achieved by the reference readers independently (κ = 0.84 ± 0.017; P = .01). Conclusion AI assistance increased junior readers' radiographic KOA grading performance and increased interobserver agreement for osteoarthritis grading across all readers and experience levels. Published under a CC BY 4.0 license. Supplemental material is available for this article.

Strategic Review of the Effectiveness of Open-Source Intelligence at the National Maritime Information Center

This research provides a qualitative analysis of having Open-Source Intelligence (OSINT) at the National Maritime Information Center (NMIC). Indonesia's geographic condition demands strong marine defense and has enormous challenges in the maritime sector, so it is necessary to integrate maritime information data systems through the National Maritime Information Center. Technological advances provide opportunities and threats related to the potential to obtain information quickly. New technologies help localize, process, and disseminate large amounts of data and prepare it for analysis coming from Open-Source Intelligence. NMIC is required to be able to meet the needs of its users in the form of short and fast intelligence reports by utilizing information from OSINT. This is what underlies researchers to conduct research with problem formulations that have the effectiveness of OSINT in the NMIC. This research uses a qualitative method with an exploratory approach to find and analyze more in-depth things that are not yet known about OSINT. Productivity theory is used by researchers to discuss problem formulations assisted by NVIVO 12 tools in processing data and Soft System Methodology (SSM) to analyze data. The results showed that to use OSINT effectively, NMIC must have large data analysis capabilities using artificial intelligence assistance.

Intelligent assistance system for people with visual impairments

Subject of the Research: The creation of an intelligent assistance system for people with visual impairments. Nowadays, the task of developing effective intelligent assistance systems that allow people with vision problems to achieve maximum independence is important and relevant, as existing systems have a number of drawbacks, such as limited autonomy, limited integration with other devices and systems, limited analysis of dynamic obstacles, and limited user feedback capabilities, which in most cases are restricted to voice guidance. Objective: The objective of this work is to create a generalized functional model of an intelligent assistance system for people with visual impairments, which has enhanced autonomy, integration with other devices and systems, the ability to analyze dynamic objects and predict their movement trajectory, and provide diverse feedback to the user. Tasks: To achieve the set objective, the following tasks were accomplished: a generalized functional model of the proposed intelligent assistance system for people with visual impairments was created; the functional dependencies of the components of the developed model were substantiated; a review of the basic modules of the proposed system model was conducted. Methods: The methods used include functional modeling methods and system analysis methods. Results: The following results were obtained: a functional model of an intelligent assistance system for people with visual impairments was proposed. This system surpasses existing analogs in a number of functional capabilities: detection of static and dynamic obstacles with prediction of dynamic obstacles’ movement trajectory, the ability to operate in various conditions (indoors, outdoors, in light or dark, in different weather conditions), support for integration with other systems and devices, and a high level of autonomy. Conclusions: The developed system model has enhanced autonomy, integration with other devices and systems, the ability to analyze dynamic objects and predict their movement trajectory, and provides diverse feedback to the user.

The Research of EEG Headset of Brain–Computer Interface for Artificial Intelligence-Related Applications

Over the past decade, a wide range of brain–computer interface (BCI) applications with artificial intelligence (AI) assistance have emerged. While using a head-mounted electroencephalogram (EEG) device for BCI application, it often requires a user to wear the device close to the scalp so that clear EEG signals can be obtained from the head directly. Those signals can be used to determine a user’s mental state. Due to the head size differences between Chinese and European users, as well as the differences between Chinese users themselves, EEG headsets with fixed-electrode-point will enable some electrodes fail to capture a user’s EEG data accurately. Combined with previous research and related design methodology, the design of a head-mounted EEG device is proposed in this paper. In this study, Chinese youth are regarded as the target user group. After evaluating different designs of device structures, the possible positions and movement methods of electrodes, an EEG headset with movable electrodes for young Chinese users is finally delivered.

AI Personal Assistant

Abstract : This paper talks about using a Voice Intelligent Assistance tool, which basically means a program that responds to your voice commands. It can help you search for things, summarize information, and set reminders, all just by talking to it. With voice recognition technology, you can access any document or file by simply saying its name. If you say a word, it will type it out for you. The tool listens to what you say, finds the right information in its database, and shows it to you. It works by first turning your spoken words into written text. You need to make sure to speak in the right language so the assistant can understand you correctly. If there’s any confusion, it will show messages on the screen to help you out. This assistant is like a software helper that does tasks based on what you tell it. It makes interacting with computer programs easier because you can just talk to them instead of typing. It can even read out summaries of reports and important information to you, making things simpler and more convenient.

Teaching Skills Cultivation of Information Technology Normal University Students from the Perspective of Group Intelligence: Research on the Training and Application Optimization of GPT Dialogue Model

With the wide application of information technology in education, the cultivation of teaching skills for information technology normal university students is particularly important. However, the traditional teaching model faces challenges, and new methods are needed to improve the teaching ability of normal university students. The purpose of this study is to explore the teaching skills training mode of information technology normal students based on the perspective of group intelligence, and to use GPT dialogue model for training and application optimization, so as to improve the teaching effect and the professional level of normal students. Through the teaching mode from the perspective of group wisdom, normal university students can have more in-depth interaction and cooperation with classmates, teachers and teaching resources, and promote the innovation and improvement of information technology teaching. Using GPT dialogue model for training and application optimization can realize personalized and intelligent teaching assistance, and provide teaching content and methods more close to the needs of students. The results show that the teaching mode based on the perspective of group intelligence and the GPT dialogue model can effectively improve the teaching skills of information technology normal university students, and provide new ideas and methods for the innovation and development in the field of education.

Impact analysis of Advanced Driver Assistance Systems (ADAS) regarding road safety – computing reduction potentials

In the present study road safety impact analysis for certain advanced driver assistance systems (ADAS) was conducted. Based on a literature review, expert interviews and current adaptations in legislation, the most promising nine ADAS were selected. The impact was analysed based on statistical crash data from Austria. Factors such as infrastructure and weather conditions, market penetration, expected functionality of sensors, user acceptance and risk homeostasis were considered. A software tool was developed to calculate the crash reduction potential of the selected ADAS for the scenarios 2025, 2030 and 2040.The results show that the ADAS related to warning/braking have the greatest future reduction potential and could lead to a reduction of up to 8,700 crashes and 70 fatalities in Austria in 2040. In addition, the Intelligent Speed Assistance system would lead to an overall crash reduction of 8% compared to current crash numbers in Austria in 2040. The Turning Assistant for heavy goods vehicles shows the lowest reduction in crashes and casualties, but due to the highest severity per crash (93 fatalities per 1,000 crashes), it nevertheless provides an important contribution to the reduction of fatalities in road traffic.However, to benefit from the ADAS safety potential, it is highly relevant that these systems are used in a correct manner. In the future, it will be necessary to provide users with more information on the correct use, benefits and limitations of the respective ADAS and to integrate the use of these systems into driver education procedures and tests.

Automatic classification and segmentation of multiclass jaw lesions in cone-beam CT using deep learning.

To develop and validate a modified deep learning (DL) model based on nnU-Net for classifying and segmenting five-class jaw lesions using cone-beam CT (CBCT). A total of 368 CBCT scans (37 168 slices) were used to train a multi-class segmentation model. The data underwent manual annotation by two oral and maxillofacial surgeons (OMSs) to serve as ground truth. Sensitivity, specificity, precision, F1-score, and accuracy were used to evaluate the classification ability of the model and doctors, with or without artificial intelligence assistance. The dice similarity coefficient (DSC), average symmetric surface distance (ASSD), and segmentation time were used to evaluate the segmentation effect of the model. The model achieved the dual task of classifying and segmenting jaw lesions in CBCT. For classification, the sensitivity, specificity, precision, and accuracy of the model were 0.871, 0.974, 0.874, and 0.891, respectively, surpassing oral and maxillofacial radiologists (OMFRs) and OMSs, approaching the specialist. With the model's assistance, the classification performance of OMFRs and OMSs improved, particularly for odontogenic keratocyst (OKC) and ameloblastoma (AM), with F1-score improvements ranging from 6.2% to 12.7%. For segmentation, the DSC was 87.2% and the ASSD was 1.359 mm. The model's average segmentation time was 40 ± 9.9 s, contrasting with 25 ± 7.2 min for OMSs. The proposed DL model accurately and efficiently classified and segmented five classes of jaw lesions using CBCT. In addition, it could assist doctors in improving classification accuracy and segmentation efficiency, particularly in distinguishing confusing lesions (eg, AM and OKC).

Vocational Education Skill Assessment and Intelligent Assistance: A Study on the Application of Machine Learning Algorithms in the Assessment of Vocational Information Literacy Teaching Ability

Investigating vocational educators' knowledge-based teaching skills across China's Vocational Education (VE) institutions, this research focuses on the practical use of Machine Learning (ML) algorithms. Instructors' efficacy must be evaluated, and this work addresses the gap. VE performs an essential role in connecting learning abilities with the demands of industry. The investigation plans on developing an adaptable, subjective assessment technique that extends within the boundaries of conventional subjective evaluation methods using modern ML techniques such as Support Vector Machines (SVM), Decision Trees (DT), and Neural Networks (NN). Each ML model's accuracy, reliability, and feasibility have been determined using data collected from 120 vocational educators encompassing various fields and regions. Researchers predict that our findings will provide perspective on how to improve vocational education settings' teaching methods and governance.

Fostering well-being: Exploring the influence of user-AI assistant relationship types on subjective well-being

As one of the essential components of daily life, artificial intelligence (AI) assistants gradually form a relationship with users as either servant or partner. This study employed a mixed-methods approach integrating both quantitative and qualitative methodologies to examine the influence of various user-AI assistant relationship types on subjective well-being. The results indicate that partnerships are more conducive to fostering subjective well-being in users than master-servant relationships. This influence is mediated through social support, which encompasses both informational support and emotional support, and is moderated by the presence of feelings of social loneliness. These findings contribute to the available literature on user-AI relationships and subjective well-being and can inform the market launch of AI assistants.

Evaluation of positioning accuracy, radiation dose and image quality: artificial intelligence based automatic versus manual positioning for CT KUB.

Recent innovations are making radiology more advanced for patient and patient services. Under the immense burden of radiology practice, Artificial Intelligence (AI) assists in obtaining Computed Tomography (CT) images with less scan time, proper patient placement, low radiation dose (RD), and improved image quality (IQ). Hence, the aim of this study was to evaluate and compare the positioning accuracy, RD, and IQ of AI-based automatic and manual positioning techniques for CT kidney ureters and bladder (CT KUB). This prospective study included 143 patients in each group who were referred for computed tomography (CT) KUB examination. Group 1 patients underwent manual positioning (MP), and group 2 patients underwent AI-based automatic positioning (AP) for CT KUB examination. The scanning protocol was kept constant for both the groups. The off-center distance, RD, and quantitative and qualitative IQ of each group were evaluated and compared. The AP group (9.66±6.361 mm) had significantly less patient off-center distance than the MP group (15.12±9.55 mm). There was a significant reduction in RD in the AP group compared with that in the MP group. The quantitative image noise (IN) was lower, with a higher signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) in the AP group than in the MP group (p<0.05). Qualitative IQ parameters such as IN, sharpness, and overall IQ also showed significant differences (p< 0.05), with higher scores in the AP group than in the MP group. The AI-based AP showed higher positioning accuracy with less off-center distance (44%), which resulted in 12% reduction in RD and improved IQ for CT KUB imaging compared with MP.

An intelligent FL-based vehicle route optimization protocol for green and sustainable IoT connected IoV

The intelligent Internet of Vehicles (IoV) provides superior results in effectively addressing complex transportation challenges. Predicting vehicle traffic, crashes, demand, location, communication, and travel safety are all critical issues in today’s transportation systems. The proposed paper optimizes vehicle traffic by incorporating reroute recommendations, increasing the use of public transportation, and providing onboard vehicle drivers with intelligent health assistance using Federated Learning (FL). This research also focuses on resolving complex transportation issues such as a vehicle’s current location, exact vehicle count information on each route, and onboard vehicle vacant seat information. Furthermore, vehicle communication contributes to the proposed system’s efficiency by avoiding communication delays or information loss to registered users and the cloud server. An intelligent FL-based scheme for vehicle route optimization has been proposed as part of this research to prevent vehicle traffic in a real-time Internet of Things (IoT) connected IoV transportation system. The vehicle detection approach determines the number of vehicles traveling on each route to recommend the best route to registered users. The effective implementation of cluster-based vehicle communication and location estimation models enhances the efficiency of the proposed system.

Humanzentrierter Gestaltungsprozess eines intelligenten Assistenzsystems in der manuellen Montage

Abstract This article presents a human-centered approach to developing an intelligent assembly assistance system, based on ISO 9241-210. Requirements were derived using user-centered methods, and a prototype was developed. A study evaluated the impact of error communication on users, highlighting challenges and potentials. The article demonstrates that the involvement of employees is crucial for the successful development of such systems.

Securing Breakneck Pace of 5G Networks Air Interfaces Through Proactive AI Monitoring

The promise of 5G networks enabling emerging technologies comes with formidable new security challenges. This paper proposes an AI-driven real-time security monitoring and incident response framework tailored to 5G infrastructure. A multi-layered architecture is presented using specialized deep learning models for radio access, edge, core, and network slice threat detection. Models including CNNs, RNNs, and transformers perform traffic analysis, signal classification, and log anomaly detection. A centralized controller aggregates model outputs into an integrated threat intelligence engine that deduces attack context and recommends mitigations. Further, a conversational bot interacts with security analysts in natural language to explain threats, suggest responses, and answer queries. The intelligent assistant is designed using dialog trees and transformer networks trained on security datasets. Evaluation on real-world 5G trial networks demonstrates 95% accuracy in classifying radio signal spoofing attacks and 98% precision in identifying malware infections. Analyst surveys confirm improved productivity and faster incident response with the AI assistant. As 5G matures, robust analytics and AI collaboration will grow increasingly critical for secure network operations. This research aims to provide both a conceptual framework and proven techniques as key enablers.

The AI humanness: how perceived personality builds trust and continuous usage intention

PurposeThe growing integration of artificial intelligence (AI) assistants and voice assistants provides a platform for AI to enter consumers’ everyday lives. As these voice assistants become ubiquitous, their widespread adoption underscores the need to understand how to create voice assistants that can naturally interact with and support users. Grounded in the stereotype content model from social psychology, this study aims to investigate the influence of perceived humanness and personality on building trust and continuous usage intentions in voice assistants. Specifically, a fresh perspective examining the determining factors that shape personality trait perceptions of competence and warmth in voice assistants is proposed.Design/methodology/approachAn online survey of 457 participants and structural equation modeling is conducted to validate the research model.FindingsAnthropomorphism, social presence and interactivity drive perceived warmth, whereas performance and effort expectations drive perceived competence. Perceived competence and perceived warmth together positively affect users’ trust in voice assistants, leading to a higher likelihood of continuous usage intentions.Originality/valueThis research provides profound theoretical contributions to the emerging field of human-AI interaction and offer practical implications for marketers aiming to leverage voice assistant personalities to build trusted and long-lasting interactions.

Budućnost trijaže

Introduction. Triage is the assessment of the patient’s condition in order to determine the urgency of treatment. It is usually performed by a nurse, often using a five-level protocol.Aim. To conduct a comparative analysis of the accuracy of categorization and diagnosis between ChatGPT (a chatbot which uses machine learning algorithms) and traditional medical triage, as well as to provide recommendations on how artificial intelligence can improve the work of medical professionals in patient triage.Methods. The literature selected for comparison is “Emergency Nursing: 5-Tier Triage Protocols”. The most common diagnoses for which patients present to the emergency department were selected for research. Then, triage categories were selected and case presentations were created. These cases were presented to ChatGPT, and its responses were comparedwith the literature.Results. ChatGPT correctly categorizes triage cases in 43.33% of cases, with an average category difference of 0.7. Although it made mistakes in 1 or 2 categories in some cases, it assigned diagnoses to a higher category for patient safety.Discussion. Comparison with other studies shows that errors occur in up to 40% of nurse decisions due to various factors such as inexperience, speed of work, and a large number of patients, which could be reduced by additional artificial intelligence assistance. It is necessary to take into account factors that artificial intelligence cannot take over and that it can only be a help, not a substitute for medical personnel.Conclusion. ChatGPT has potential for usage in medical triage, but with improvements in specialized training of models on medical data and terminology to improve the accuracy and reliability of the model.

Analytical modeling of perceived authenticity in AI assistants: application of PLS-predict algorithm and importance-performance map analysis

PurposeThe present study aimed to explore the consequences of perceived authenticity in artificial intelligence (AI) assistants and develop a serial-mediation architecture specifying causation of loyalty in human–AI relationships. It intended to assess the predictive power of the developed model based on a training-holdout sample procedure. It further attempted to map and examine the predictors of loyalty, strengthening such relationship.Design/methodology/approachPartial least squares structural equation modeling (PLS-SEM) based on bootstrapping technique was employed to examine the higher-order effects pertaining to human–AI relational intricacies. The sample size of the study comprised of 412 AI assistant users belonging to millennial generation. PLS-Predict algorithm was used to assess the predictive power of the model, while importance-performance analysis was executed to assess the effectiveness of the predictor variables on a two-dimensional map.FindingsA positive relationship was found between “Perceived Authenticity” and “Loyalty,” which was serially mediated by “Perceived-Quality” and “Animacy” in human–AI relational context. The construct “Loyalty” remained a significant predictor of “Emotional-Attachment” and “Word-of-Mouth.” The model possessed high predictive power. Mapping analysis delivered contradictory result, indicating “authenticity” as the most significant predictor of “loyalty,” but the least effective on performance dimension.Practical implicationsThe findings of the study may assist marketers to understand the relevance of AI authenticity and examine the critical behavioral consequences underlying customer retention and extension strategies.Originality/valueThe study is pioneer to introduce a hybrid AI authenticity model and establish its predictive power in explaining the transactional and communal view of human reciprocation in human–AI relationship. It exclusively provided relative assessment of the predictors of loyalty on a two-dimensional map.

Importance Performance Matrix Analysis for Assessing User Experience with Intelligent Voice Assistants: A Strategic Evaluation

Importance Performance Matrix Analysis for Assessing User Experience with Intelligent Voice Assistants: A Strategic Evaluation