Artificial intelligence-assisted reduction in patients' waiting time for outpatient process: a retrospective cohort study

Constructing and implementing a performance evaluation indicator set for artificial intelligence decision support systems in pediatric outpatient clinics: an observational study

Artificial intelligence (AI) decision support systems in pediatric healthcare have a complex application background. As an AI decision support system (AI-DSS) can be costly, once applied, it is crucial to focus on its performance, interpret its success, and then monitor and update it to ensure ongoing success consistently. Therefore, a set of evaluation indicators was explicitly developed for AI-DSS in pediatric healthcare, enabling continuous and systematic performance monitoring. The study unfolded in two stages. The first stage encompassed establishing the evaluation indicator set through a literature review, a focus group interview, and expert consultation using the Delphi method. In the second stage, weight analysis was conducted. Subjective weights were calculated based on expert opinions through analytic hierarchy process, while objective weights were determined using the entropy weight method. Subsequently, subject and object weights were synthesized to form the combined weight. In the two rounds of expert consultation, the authority coefficients were 0.834 and 0.846, Kendall's coordination coefficient was 0.135 in Round 1 and 0.312 in Round 2. The final evaluation indicator set has three first-class indicators, fifteen second-class indicators, and forty-seven third-class indicators. Indicator I-1(Organizational performance) carries the highest weight, followed by Indicator I-2(Societal performance) and Indicator I-3(User experience performance) in the objective and combined weights. Conversely, 'Societal performance' holds the most weight among the subjective weights, followed by 'Organizational performance' and 'User experience performance'. In this study, a comprehensive and specialized set of evaluation indicators for the AI-DSS in the pediatric outpatient clinic was established, and then implemented. Continuous evaluation still requires long-term data collection to optimize the weight proportions of the established indicators.

Applications of Artificial Intelligence in Emergency Departments to Improve Wait Times: Protocol for an Integrative Living Review

BACKGROUND

Long wait times in the emergency department (ED) are a major issue for health care systems all over the world. The application of artificial intelligence (AI) is a novel strategy to reduce ED wait times when compared to the interventions included in previous research endeavors. To date, comprehensive systematic reviews that include studies involving AI applications in the context of EDs have covered a wide range of AI implementation issues. However, the lack of an iterative update strategy limits the use of these reviews. Since the subject of AI development is cutting edge and is continuously changing, reviews in this area must be frequently updated to remain relevant.

OBJECTIVE

This study aims to provide a summary of the evidence that is currently available regarding how AI can affect ED wait times; discuss the applications of AI in improving wait times; and periodically assess the depth, breadth, and quality of the evidence supporting the application of AI in reducing ED wait times.

METHODS

We plan to conduct a living systematic review (LSR). Our strategy involves conducting continuous monitoring of evidence, with biannual search updates and annual review updates. Upon completing the initial round of the review, we will refine the search strategy and establish clear schedules for updating the LSR. An interpretive synthesis using Whittemore and Knafl's framework will be performed to compile and summarize the findings. The review will be carried out using an integrated knowledge translation strategy, and knowledge users will be involved at all stages of the review to guarantee applicability, usability, and clarity of purpose.

RESULTS

The literature search was completed by September 22, 2023, and identified 17,569 articles. The title and abstract screening were completed by December 9, 2023. In total, 70 papers were eligible. The full-text screening is in progress.

CONCLUSIONS

The review will summarize AI applications that improve ED wait time. The LSR enables researchers to maintain high methodological rigor while enhancing the timeliness, applicability, and value of the review.

INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID)

DERR1-10.2196/52612.

Artificial intelligence for detection of effusion and lipo-hemarthrosis in X-rays and CT of the knee

BACKGROUND

Traumatic knee injuries are challenging to diagnose accurately through radiography and to a lesser extent, through CT, with fractures sometimes overlooked. Ancillary signs like joint effusion or lipo-hemarthrosis are indicative of fractures, suggesting the need for further imaging. Artificial Intelligence (AI) can automate image analysis, improving diagnostic accuracy and help prioritizing clinically important X-ray or CT studies.

OBJECTIVE

To develop and evaluate an AI algorithm for detecting effusion of any kind in knee X-rays and selected CT images and distinguishing between simple effusion and lipo-hemarthrosis indicative of intra-articular fractures.

METHODS

This retrospective study analyzed post traumatic knee imaging from January 2016 to February 2023, categorizing images into lipo-hemarthrosis, simple effusion, or normal. It utilized the FishNet-150 algorithm for image classification, with class activation maps highlighting decision-influential regions. The AI's diagnostic accuracy was validated against a gold standard, based on the evaluations made by a radiologist with at least four years of experience.

RESULTS

Analysis included CT images from 515 patients and X-rays from 637 post traumatic patients, identifying lipo-hemarthrosis, simple effusion, and normal findings. The AI showed an AUC of 0.81 for detecting any effusion, 0.78 for simple effusion, and 0.83 for lipo-hemarthrosis in X-rays; and 0.89, 0.89, and 0.91, respectively, in CTs.

CONCLUSION

The AI algorithm effectively detects knee effusion and differentiates between simple effusion and lipo-hemarthrosis in post-traumatic patients for both X-rays and selected CT images further studies are needed to validate these results.

The Role of AI in Hospitals and Clinics: Transforming Healthcare in the 21st Century

As healthcare systems around the world face challenges such as escalating costs, limited access, and growing demand for personalized care, artificial intelligence (AI) is emerging as a key force for transformation. This review is motivated by the urgent need to harness AI's potential to mitigate these issues and aims to critically assess AI's integration in different healthcare domains. We explore how AI empowers clinical decision-making, optimizes hospital operation and management, refines medical image analysis, and revolutionizes patient care and monitoring through AI-powered wearables. Through several case studies, we review how AI has transformed specific healthcare domains and discuss the remaining challenges and possible solutions. Additionally, we will discuss methodologies for assessing AI healthcare solutions, ethical challenges of AI deployment, and the importance of data privacy and bias mitigation for responsible technology use. By presenting a critical assessment of AI's transformative potential, this review equips researchers with a deeper understanding of AI's current and future impact on healthcare. It encourages an interdisciplinary dialogue between researchers, clinicians, and technologists to navigate the complexities of AI implementation, fostering the development of AI-driven solutions that prioritize ethical standards, equity, and a patient-centered approach.

Acceptance of Medical Artificial Intelligence in Skin Cancer Screening: Choice-Based Conjoint Survey

BACKGROUND

There is great interest in using artificial intelligence (AI) to screen for skin cancer. This is fueled by a rising incidence of skin cancer and an increasing scarcity of trained dermatologists. AI systems capable of identifying melanoma could save lives, enable immediate access to screenings, and reduce unnecessary care and health care costs. While such AI-based systems are useful from a public health perspective, past research has shown that individual patients are very hesitant about being examined by an AI system.

OBJECTIVE

The aim of this study was two-fold: (1) to determine the relative importance of the provider (in-person physician, physician via teledermatology, AI, personalized AI), costs of screening (free, 10€, 25€, 40€; 1€=US $1.09), and waiting time (immediate, 1 day, 1 week, 4 weeks) as attributes contributing to patients' choices of a particular mode of skin cancer screening; and (2) to investigate whether sociodemographic characteristics, especially age, were systematically related to participants' individual choices.

METHODS

A choice-based conjoint analysis was used to examine the acceptance of medical AI for a skin cancer screening from the patient's perspective. Participants responded to 12 choice sets, each containing three screening variants, where each variant was described through the attributes of provider, costs, and waiting time. Furthermore, the impacts of sociodemographic characteristics (age, gender, income, job status, and educational background) on the choices were assessed.

RESULTS

Among the 383 clicks on the survey link, a total of 126 (32.9%) respondents completed the online survey. The conjoint analysis showed that the three attributes had more or less equal importance in contributing to the participants' choices, with provider being the most important attribute. Inspecting the individual part-worths of conjoint attributes showed that treatment by a physician was the most preferred modality, followed by electronic consultation with a physician and personalized AI; the lowest scores were found for the three AI levels. Concerning the relationship between sociodemographic characteristics and relative importance, only age showed a significant positive association to the importance of the attribute provider (r=0.21, P=.02), in which younger participants put less importance on the provider than older participants. All other correlations were not significant.

CONCLUSIONS

This study adds to the growing body of research using choice-based experiments to investigate the acceptance of AI in health contexts. Future studies are needed to explore the reasons why AI is accepted or rejected and whether sociodemographic characteristics are associated with this decision.

Improving radiology workflow using ChatGPT and artificial intelligence

Artificial Intelligence is a branch of computer science that aims to create intelligent machines capable of performing tasks that typically require human intelligence. One of the branches of artificial intelligence is natural language processing, which is dedicated to studying the interaction between computers and human language. ChatGPT is a sophisticated natural language processing tool that can understand and respond to complex questions and commands in natural language. Radiology is a vital aspect of modern medicine that involves the use of imaging technologies to diagnose and treat medical conditions artificial intelligence, including ChatGPT, can be integrated into radiology workflows to improve efficiency, accuracy, and patient care. ChatGPT can streamline various radiology workflow steps, including patient registration, scheduling, patient check-in, image acquisition, interpretation, and reporting. While ChatGPT has the potential to transform radiology workflows, there are limitations to the technology that must be addressed, such as the potential for bias in artificial intelligence algorithms and ethical concerns. As technology continues to advance, ChatGPT is likely to become an increasingly important tool in the field of radiology, and in healthcare more broadly.

Effect of waiting time on patient satisfaction in outpatient: An empirical investigation

This study aimed to identify the effect of waiting time on patient satisfaction and the relationship between different types of waiting time. The questionnaire contained 2 parts. The first part included questions about expected waiting time (EWT), reasonable waiting time, tolerance waiting time, and basic personal information. The second part included perceived waiting time (PWT) and satisfaction evaluation. The actual waiting time (AWT) was recorded by the worker. Linear regression was used to analyze the influence of waiting time on satisfaction. Before data collection, this study was approved by the hospital's health ethics committee. In total, 323 questionnaires were collected, of which 292 (90.4%) were valid. The EWT, tolerance waiting time, rational waiting time, and PWT had a significant effect on patient satisfaction (P = .006, P = .043, P = .009, P = .000), whereas AWT had no significant effect on satisfaction (P = .365). The difference between the EWT and AWT had a significant effect on satisfaction (P = .000), while the difference between the PWT and AWT had a significant effect on satisfaction (P = .000). Age, educational background, gender, appointment, and hospital visit experience had no significant effect on patient satisfaction (P = .105, P = .443, P = .260, P = .352, P = .461, respectively). Patient satisfaction with waiting time was not directly affected by AWT, but by subjective waiting times. Furthermore, objective waiting time affects patient satisfaction through the subjective waiting time. Therefore, hospital managers can improve service quality by focusing on adjusting a patient's subjective waiting time while reducing the objective waiting time.

How to adjust the expected waiting time to improve patient's satisfaction?

BACKGROUND

Long waiting time in hospital leads to patient's low satisfaction. In addition to reducing the actual waiting time (AWT), we can also improve satisfaction by adjusting the expected waiting time (EWT). Then how much can the EWT be adjusted to attribute a higher satisfaction?

METHODS

This study was conducted though experimental with hypothetical scenarios. A total of 303 patients who were treated by the same doctor from August 2021 to April 2022 voluntarily participated in this study. The patients were randomly divided into six groups: a control group (n = 52) and five experimental groups (n = 245). In the control group, the patients were asked their satisfaction degree regarding a communicated EWT (T₀) and AWT (T_a) under a hypothetical situation. In the experimental groups, in addition to the same T₀ and T_a as the control group, the patients were also asked about their satisfaction degree with the extended communicated EWT (T₁). Patients in five experimental groups were given T₁ values with 70, 80, 90, 100, and 110 min respectively. Patients in both control and experiment groups were asked to indicate their initial EWT, after given unfavorable information (UI) in a hypothetical situation, the experiment groups were asked to indicate their extended EWT. Each participant only participated in filling out one hypothetical scenario. 297 valid hypothetical scenarios were obtained from the 303 hypothetical scenarios given.

RESULTS

The experimental groups had significant differences between the initial indicated EWT and extended indicated EWT under the effect of UI (20 [10, 30] vs. 30 [10, 50], Z = -4.086, P < 0.001). There was no significant difference in gender, age, education level and hospital visit history (χ² = 3.198, P = 0.270; χ² = 2.177, P = 0.903; χ² = 3.988, P = 0.678; χ² = 3.979, P = 0.264) in extended indicated EWT. As for patient's satisfaction, compared with the control group, significant differences were found when T₁ = 80 min (χ² = 13.511, P = 0.004), T₁ = 90 min (χ² = 12.207, P = 0.007) and T₁ = 100 min (χ² = 12.941, P = 0.005). When T₁ = 90 min, which is equal to the T_a, 69.4% (34/49) of the patients felt "very satisfied", this proportion is not only significantly higher than that of the control group (34/ 49 vs. 19/52, χ² = 10.916, P = 0.001), but also the highest among all groups. When T₁ = 100 min (10 min longer than T_a), 62.5% (30/48) of the patients felt "very satisfied", it is significantly higher than that of the control group (30/ 48 vs. 19/52, χ² = 6.732, P = 0.009). When T₁ = 80 min (10 min shorter than T_a), 64.8% (35/54) of the patients felt "satisfied", it is significantly higher than that of the control group (35/ 54 vs. 17/52, χ² = 10.938, P = 0.001). However, no significant difference was found when T₁ = 70 min (χ² = 7.747, P = 0.052) and T₁ = 110 min (χ² = 4.382, P = 0.223).

CONCLUSIONS

Providing UI prompts can extend the EWT. When the extended EWT is closer to the AWT, the patient's satisfaction level can be improved higher. Therefore, medical institutions can adjust the EWT of patient's through UI release according to the AWT of hospitals to improve patient's satisfaction.

Stature estimation by semi-automatic measurements of 3D CT images of the femur

Stature estimation is one of the most basic and important methods of personal identification. The long bones of the limbs provide the most accurate stature estimation, with the femur being one of the most useful. In all the previously reported methods of stature estimation using computed tomography (CT) images of the femur, laborious manual measurement was necessary. A semi-automatic bone measuring method can simplify this process, so we firstly reported a stature estimation process using semi-automatic bone measurement software equipped with artificial intelligence. Multiple measurements of femurs of adult Japanese cadavers were performed using automatic three-dimensional reconstructed CT images of femurs. After manually setting four points on the femur, an automatic measurement was acquired. The relationships between stature and five femoral measurements, with acceptable intraobserver and interobserver errors, were analyzed with single regression analysis using the standard error of the estimate (SEE) and the coefficient of determination (R²). The maximum length of the femur (MLF) provided the lowest SEE and the highest R²; the SEE and R² in all cadavers, males and females, respectively, were 3.913 cm (R² = 0.842), 3.664 cm (R² = 0.705), and 3.456 cm (R² = 0.686) for MLF on the right femur, and 3.837 cm (R² = 0.848), 3.667 cm (R² = 0.705), and 3.384 cm (R² = 0.699) for MLF on the left femur. These results were non-inferior to those of previous reports regarding stature estimation using the MLF. Stature estimation with this simple and time-saving method would be useful in forensic medical practice.

The Association between mHealth App Use and Healthcare Satisfaction among Clients at Outpatient Clinics: A Cross-Sectional Study in Inner Mongolia, China

Mobile health (mHealth) applications (apps) have been developed in hospital settings to allocate and manage medical care services, which is one of the national strategies to improve health care in China. Little is known about the comprehensive effects of hospital-based mHealth app use on client satisfaction. The aim of this study was to determine the relationship between the full range of mHealth app use and satisfaction domains among clients attending outpatient clinics. A cross-sectional survey was conducted from January to February 2021 in twelve tertiary hospitals in Inner Mongolia. After the construction of the mHealth app use, structural equation modeling was used for data analysis. Of 1889 participants, the standardized coefficients β on environment/convenience, health information, and medical service fees were 0.11 (p < 0.001), 0.06 (p = 0.039), and 0.08 (p = 0.004), respectively. However, app use was not significantly associated with satisfaction of doctor−patient communication (β = 0.05, p = 0.069), short-term outcomes (β = 0.05, p = 0.054), and general satisfaction (β = 0.02, p = 0.429). Clients of the study hospitals were satisfied with the services, but their satisfaction was not much associated with mHealth use. The limitation of the mHealth system should be improved to enhance communication and engagement among clients, doctors, and healthcare givers, as well as to pay more attention to health outcomes and satisfaction of clients.

Using artificial intelligence to reduce queuing time and improve satisfaction in pediatric outpatient service: A randomized clinical trial

INTRODUCTION

Complicated outpatient procedures are associated with excessive paperwork and long waiting times. We aimed to shorten queuing times and improve visiting satisfaction.

METHODS

We developed an artificial intelligence (AI)-assisted program named Smart-doctor. A randomized controlled trial was conducted at Shanghai Children's Medical Center. Participants were randomly divided into an AI-assisted and conventional group. Smart-doctor was used as a medical assistant in the AI-assisted group. At the end of the visit, an e-medical satisfaction questionnaire was asked to be done. The primary outcome was the queuing time, while secondary outcomes included the consulting time, test time, total time, and satisfaction score. Wilcoxon rank sum test, multiple linear regression and ordinal regression were also used.

RESULTS

We enrolled 740 eligible patients (114 withdrew, response rate: 84.59%). The median queuing time was 8.78 (interquartile range [IQR] 3.97,33.88) minutes for the AI-assisted group versus 21.81 (IQR 6.66,73.10) minutes for the conventional group (p < 0.01), and the AI-assisted group had a shorter consulting time (0.35 [IQR 0.18, 0.99] vs. 2.68 [IQR 1.82, 3.80] minutes, p < 0.01), and total time (40.20 [IQR 26.40, 73.80] vs. 110.40 [IQR 68.40, 164.40] minutes, p < 0.01). The overall satisfaction score was increased by 17.53% (p < 0.01) in the AI-assisted group. In addition, multiple linear regression and ordinal regression showed that the queuing time and satisfaction were mainly affected by group (p < 0.01), and missing the turn (p < 0.01).

CONCLUSIONS

Using AI to simplify the outpatient service procedure can shorten the queuing time of patients and improve visit satisfaction.

Prediction across healthcare settings: a case study in predicting emergency department disposition

Several approaches exist today for developing predictive models across multiple clinical sites, yet there is a lack of comparative data on their performance, especially within the context of EHR-based prediction models. We set out to provide a framework for prediction across healthcare settings. As a case study, we examined an ED disposition prediction model across three geographically and demographically diverse sites. We conducted a 1-year retrospective study, including all visits in which the outcome was either discharge-to-home or hospitalization. Four modeling approaches were compared: a ready-made model trained at one site and validated at other sites, a centralized uniform model incorporating data from all sites, multiple site-specific models, and a hybrid approach of a ready-made model re-calibrated using site-specific data. Predictions were performed using XGBoost. The study included 288,962 visits with an overall admission rate of 16.8% (7.9–26.9%). Some risk factors for admission were prominent across all sites (e.g., high-acuity triage emergency severity index score, high prior admissions rate), while others were prominent at only some sites (multiple lab tests ordered at the pediatric sites, early use of ECG at the adult site). The XGBoost model achieved its best performance using the uniform and site-specific approaches (AUC = 0.9–0.93), followed by the calibrated-model approach (AUC = 0.87–0.92), and the ready-made approach (AUC = 0.62–0.85). Our results show that site-specific customization is a key driver of predictive model performance.