Clinical Research Directory

Ambient AI Clinical Trial

This is a single-site pragmatic randomized control trial studying the effect of ambient artificial intelligence (AI) scribes on the delivery of medical care to patients in the ambulatory setting. The study will last 150 days and include up to 65 providers in the intervention group. Providers will be recruited from three medical specialties, including primary care, oncology, and urology. The study will enroll providers and randomize them to an intervention group (access to the ambient AI scribe product) or a control group (routine patient care). Providers will be evaluated for burnout and task load measures through digital surveys at the beginning, middle, and end of the study. Provider electronic health record (EHR) usage data will also be evaluated for time spent documenting, time spent after hours on days with scheduled clinical care, and time between the start of the clinical encounter and signing it.

Effect of AI-Supported Case Analysis on Nursing Students

The aim of this study is to determine the effect of AI-supported oncology case analysis on nursing students' knowledge, level of learning satisfaction, and clinical decision-making skills. This study is planned to be conducted using a single-blind randomized controlled trial design for the quantitative research component and an interview design for the qualitative research component. The students will be divided into two groups: an intervention group (artificial intelligence) and a control group (traditional instruction).

Non-Contrast Breast MRI Diagnosis and Risk Stratification Using DWI-Generated Synthetic Contrast Enhancement

The goal of this observational study is to develop an integrated breast MRI system that uses diffusion-weighted imaging (DWI) to create synthetic contrast-enhanced images. This system aims to diagnose and screen for breast cancer without the need for contrast agents, while using a generated risk score to perform imaging-based triage and risk stratification. Participants will include people aged 18 and older who require a breast MRI either for evaluation of a suspicious finding or for high-risk screening. This study seeks to answer two main questions: * Can synthetic contrast-enhanced images generated from DWI match real contrast-enhanced images in their ability to distinguish benign from malignant breast lesions? * Can the risk score derived from DWI-based synthetic images enable imaging-level risk stratification, allowing people at lower risk to avoid contrast agent injection? Researchers will compare the quality of synthetic images against real contrast-enhanced images and will recruit radiologists to assess how well these images perform for diagnostic and screening tasks. MRI data from participants undergoing breast MRI will be used to train, validate, and test this integrated system.

Development of a Chatbot-supported Personalized Exercise Program for Older Adults and Evaluation of Its Effects on Cognitive Functions

The purpose of this study is to develop an artificial intelligence-based chatbot application to support exercise behavior in individuals aged 60 and over who do not regularly exercise, and to evaluate its effectiveness. In addition, the study aims to examine the effects of changes in exercise habits on the cognitive (mental) functions of older adults. In this study, the impact of a chatbot-supported personalized exercise program on cognitive functions in older individuals will be evaluated. A total of 90 participants is planned for inclusion in this study. If you agree to participate in this study, depending on the group you are assigned to, you may receive: * An artificial intelligence-based chatbot program, along with educational materials about the importance of exercise, or * Only educational materials (brochures) prepared by the researchers about the importance of exercise. At the beginning of the study, you will be asked to complete a data collection form. The same form will also be administered at week 12 and week 24. This form will include: * Basic information such as your age and gender, * Questions about your exercise habits, * A brief test to assess your cognitive (mental) functions, * Questions evaluating your level of physical activity. The study duration is 24 weeks, including 12 weeks of intervention and 12 weeks of follow-up.

AI-GF-GNW on Prolonged Grief Reactions

Prolonged Grief Disorder (PGD) is a severe, disabling condition characterized by intense yearning and difficulty accepting the reality of loss, which significantly impairs the academic and psychosocial functioning of bereaved adolescents. While Grief-Focused Cognitive Behavioral Therapy (GF-CBT) is effective, its high cost and resource-intensive nature limit its accessibility for adolescents in mainland China. Grief-Focused Guided Narrative Writing (GF-GNW) offers a scalable, low-cost alternative that facilitates memory integration. Furthermore, integrating Artificial Intelligence (AI) to provide personalized, structured feedback has the potential to simulate therapist functions and enhance intervention efficacy. However, the specific efficacy of AI-assisted feedback in this context remains empirically unvalidated. This parallel randomized controlled trial aims to examine the effectiveness of AI-assisted GF-GNW (AI-GF-GNW) in treating Chinese adolescents (aged 10-19) with subclinical PGD, compared to a no-feedback NF-GF-GNW group and a free writing group. Primary outcomes include PGD symptom severity, while secondary outcomes assess depression, anxiety, and daily functioning. We hypothesize that both active intervention arms will significantly alleviate PGD and related symptoms compared to the free writing group, and that the AI-GF-GNW group will demonstrate a significantly greater reduction in symptoms and functional impairment than the NF-GF-GNW group.

Effectiveness of Large Language Model for Anaesthesia and Procedural Consent

Patient understanding of anaesthesia risks remains inconsistent due to time constraints, language barriers, and variable clinician communication styles. Traditional verbal consent may not consistently ensure comprehension or reduce preoperative anxiety. PEAR (Patient Education of Anesthesia Risks) is a multilingual, AI-driven chatbot developed to enhance patient education and improve the quality of anaesthesia risk counselling. Study Objective: To compare PEAR's performance in delivering anaesthesia risk consent against the standard face-to-face verbal method.

Identification of Depressive and Anxiety Symptoms Among a Sample of Emergency Department Patients Using Artificial Intelligence (AI) Technology

Behavioral health problems, such as depression and anxiety, are common yet often are not identified by emergency department doctors and nurses. These mental health conditions can be due to medical issues or can worsen medical problems. One way investigators hope to do a better job of learning about mental health is by training Artificial Intelligence (AI) software to detect anxiety and depression by analyzing facial expression and tone of voice. Participants are invited to participate in a study which may help improve emergency department care. An audio and video recording of the participant's responses to some simple, non-psychological questions will be analyzed by a computer to determine whether investigators can assess mood and anxiety by analyzing speech and visual patterns. The audio and video will not be listened to nor watched by study personnel, only analyzed by a computer. The investigator's hope is that it will help others in the future by aiding in the assessment of psychological state. This study is being conducted at CMC ED only.

A Randomized Controlled Trial of Ambient Artificial Intelligence Scribe Technologies

This is a three-arm pragmatic RCT of 238 outpatient physicians at a large academic health system, randomized 1:1:1 to one of two AI scribe tools or a usual-care control group. The two-month study will observe and compare the effects of each tool prior to system-wide roll out of selected tool (anticipated Spring 2025). We will use covariate-constrained randomization to balance the arms in terms of physician baseline time in notes, survey-measured level of burnout, and clinic days per week. The primary purpose of the initiative is to improve quality, efficiency, and business operations at University of California, Los Angeles (UCLA) Health, and this initiative is not being done for research purposes. The results of this operational initiative will inform the widespread roll out of AI scribe tools across all providers within the UCLA Health System. Nevertheless, the UCLA study team plans to rigorously examine and publish the impact of this intervention across the health system, which is why the study team pre-registered the initiative.

A Prospective Study to Evaluate the Performance of a Real-time System in the Estimation of Colorectal Polyp Size

Colorectal polyp size is related to the risk of exhibiting advanced histological features. Moreover, polyps larger than 10 mm are associated with an elevated risk of metachronous advanced neoplasia and colorectal cancer (CRC). Consequently, accurate measurement of polyp size, especially at the 10 mm threshold is critical for risk stratification and surveillance intervals. Furthermore, polyp size is also important for the choice of the appropriate resection procedures. Underestimation may lead to delayed diagnosis, thereby increasing the risk of colorectal cancer, while overestimation may result in unnecessary surveillance endoscopies.

Artificial Intelligence Education and Climate Awareness in Pregnancy

This quasi-experimental pretest-posttest study aimed to evaluate the effect of artificial intelligence-assisted climate change education on pregnant women's climate change concerns and awareness. The study will be conducted with pregnant women attending a pregnancy school, and participants will be assigned to intervention and control groups. The intervention group will receive AI-supported climate change education in addition to routine training, while the control group will receive only routine education. Data will be collected using the Climate Change Anxiety Scale and the Maternal-Fetal Health Awareness of Climate Change Scale. The findings are expected to contribute to improving pregnant women's awareness and reducing concerns related to climate change through innovative educational approaches.

Structured Handoff Using Intelligent Framework for Transitions Trial

Inpatient general medicine attendings will be randomized to have an LLM feature turned on to provide a draft of an off-service handoff within Carelign (an EHR-adjacent provider communication tool). Providers who have access to this feature will be clearly instructed that if they use the LLM-generated draft, they must review and edit it as necessary before finalizing. The study will assess measures of documentation burden (as it relates to writing handoff) - including time spent writing handoff - and work exhaustion in both intervention and control groups.

AMIE's Clinical Conversational Abilities in an Urgent Care Setting

The purpose of this study is to determine the feasibility of a conversational artificial intelligence (AI) system to have a meaningful clinical conversation with a patient prior to an urgent care visit with their primary care physician. In this study, patients who are seeking an urgent care visit (that is, any type of medical visit with their primary care provider for a new complaint) will first have a conversation with an AI system. This interaction with the AI system will happen less than a week before their visit with their physician, and will be supervised by an independent physician who will interrupt in case there are any concerns about patient safety. After the interaction, a summary of the conversation will be sent to the patient's PCP, who will review prior to the in-person visit. The researchers will investigate: * Patient views on the AI system * PCP views on the AI system * Overall safety, as measured by the physician safety supervisor * Quality of clinical conversations, measured by standardized rubrics * Quality of diagnostic and management plans generated by the AI; these will not be shared with the patient or physician, but will be generated after the fact and compared with the actual diagnosis and management plan.

Barriers and Facilitators to Nursing Record With AI Technology Application

The goal of this study is to examine the facilitators and barriers to the comprehensive implementation of AI technology in nursing documentation. The main questions it aims to answer are: What are facilitators to the comprehensive implementation of AI technology in nursing documentation? What are barriers to the comprehensive implementation of AI technology in nursing documentation? What strategies can help to fully utilize artificial intelligence technology in nursing documentation?

Assessing Intensive Care Unit (ICU) Indications: Human vs. ChatGPT-4o Predictions

This retrospective study evaluates the accuracy of ICU admission indications by comparing clinical decisions with predictions from ChatGPT-4. Patient data, including demographics, vital signs, laboratory results, imaging findings, and clinical decisions, will be retrospectively collected and documented systematically using Case Report Forms. The model will be trained using ICU admission guidelines and tasked to predict ICU needs based on collected patient data. This study aims to systematically assess the alignment between AI-based predictions and clinical decisions for ICU admissions.

The Impact of Artificial Intelligence Electrocardiography on Occlusion Myocardial Infarction Management Under the Value-Based Payment System

This trial will prospectively evaluate the impact of integrating AI-ECG within the pay-for-performance program on improving the diagnosis, treatment, and clinical outcomes of occlusion myocardial infarction patients by promoting accurate and timely diagnoses through financial incentives.

Comparing Original Patient Educational Materials vs. AI-Simplified Materials to Improve Patient Comprehension and Health Literacy

Poor health literacy and patient comprehension have been associated with adverse health outcomes. Patient educational materials (PEMs) are articles that are intended to assist patients in their understanding of a given medical condition. Given that the average American adult reads at the 8th grade level, the American Medical Association and the Center for Disease Control recommend PEM be written at the 6th grade level. However, literature has found the majority of PEMs to be written significantly higher than the 8th grade level. In order to improve their readability, a number of studies have displayed the effectiveness of large language models (LLMs) such as ChatGPT to simplify the text of a given PEM. Despite the improvement in readability, the effectiveness of these simplified PEMs on improving patient comprehension of the AI augmented material has yet to be investigated. The purpose of our study is to test whether the improvement in readability found in AI-simplified PEMs corresponds to a greater understanding of the material compared to the original PEM. Understanding if AI-simplified PEM truly improves comprehension could further support this use case for AI and aid providers and healthcare organizations in improving the health literacy of their patients. This study aims to answer the following question: Do AI simplified PEMs improve the comprehension of pediatric orthopaedic conditions? Researchers will compare AI-simplified PEMs to their original, unmodified counterparts in order to see if there is any difference in post reading comprehension of the participants. Participation in the study will include: * A brief baseline survey (e.g. demographics and educational attainment) * A randomly assigned reading of either the original PEM or the AI simplified version. * A 10 question post-reading multiple choice quiz

Screening Mammography: Single Reading by One Radiologist With AI vs. Double Reading by Two Radiologists (AI-BCSQ)

A randomized prospective study comparing the evaluation of mammography images in a breast cancer screening programme by a single radiologist with AI support versus standard double reading by two radiologists without AI support.

Application Evaluation Research on the Artificial Intelligence-assisted Support System for the Diagnosis of Colorectal Tubular Adenoma Lesions

This study is a prospective,multi-center and observational clinical study.Investigators would like to innovatively construct a "trinity" database of colorectal tubular adenomas based on white light - magnifying chromo - pathological images.It simulates the decision - making logic of doctors, and based on the multimodal endoscopic LAFEQ method previously proposed, develop a multimodal deep - learning diagnostic model for colon adenomas and an interpretable risk prediction model for intestinal adenomas. While achieving high - precision auxiliary treatment decisions, clearly present the decision - making basis, and break through the limitation of poor interpretability of previous medical imaging AI models.

Lymphoedema Diagnosis and Treatment

A domain-specific, custom-trained large language model for the differential diagnosis and treatment planning of lymphedema, lipedema, and venous insufficiency.

Behavioral Parent Training With and Without AI Support for Children With Disruptive Behaviors

The goal of this clinical trial is to learn if adding an artificial intelligence (AI) application called to standard Behavioral Parent Training (BPT) helps families with children who have disruptive behavior problems. It will also help researchers understand if the app is easy to use and helpful for parents. The main question it aims to answer is: \- Is it feasible and acceptable for parents to use the AI app alongside their therapy sessions? The secondary questions it aims to answer are: * Does the app help reduce children's disruptive behaviors and irritability more than therapy alone? * Does using the app help lower stress, anxiety, and depression levels for the parents? Researchers will compare: 1. Standard BPT: Parents receive 8 weekly group training sessions (online). 2. BPT plus ParenteAI: Parents receive the same 8 weekly sessions plus 24/7 access to an AI virtual assistant for personalized support. Participants will: * Attend 8 weekly group training sessions. * Complete surveys about their child's behavior and their own well-being at baseline, after group training sessions 4 and 8, and 3 and 6 months after finalizing the group training. * If in the experimental group, use the ParenteAI app to get real-time coaching and support for managing their child's behavior at home. * Provide feedback on their experience and satisfaction with the program.

Developing a Multimodal Cancer Pain Database to Support AI-Based Automatic Pain Assessment

The goal of this observational study is to collect short video and sound recordings of people with cancer to create a secure database that can be used in future research to develop an artificial intelligence (AI) tool for pain assessment. The main aim is to build a large, high-quality collection of audiovisual data showing how people with cancer express themselves when they do and do not have pain. Participants will include adults with cancer who are admitted to the oncology ward for pain treatment and a control group admitted for chemotherapy who have no pain. After giving consent, participants will: * Be recorded on video (from the shoulders up) for up to 60 seconds while reading a short sentence and describing their pain or daily experience. * Complete a short questionnaire about their mood and pain expression. * Allow researchers to collect some information from their medical record, such as their pain score, medications, and cancer type. These recordings will be securely stored and used to create a database for future AI research. No medical tests, new treatments, or extra hospital visits are involved. This study will provide the foundation for developing future AI-based tools that could support doctors and patients in monitoring and managing pain more accurately and easily.

AI-Assisted Analgesia Copilot System

The primary objective of the SEASCAPE project is to design, develop, and to apply a clinical implementation tool of a machine learning (ML) and artificial intelligence (AI)-based co-pilot system for the real-time monitoring and control of nociception during general anesthesia (GA). The ultimate clinical purpose is to optimize individualized pain management by achieving precise titration of intravenous opioids (specifically remifentanil), thereby minimizing the incidence of over- and under-dosing. This optimization is projected to enhance patient outcomes, reduce opioid-related complications, and improve overall cost-effectiveness of anesthetic procedures. The main scientific question guiding this work is: Can a novel algorithm be generated and validated to provide superior analytical precision for analgesic management by reliably differentiating genuine nociceptive responses from confounding physiological variables-such as inadequate neuromuscular blockade or changes in depth of anesthesia-thereby significantly improving the clinical decision-making framework for intraoperative nociception control? This project addresses the recognized challenge in anesthesiology: defining an objective measure to quantify nociception and antinociception during GA. Study Population: Patients scheduled for elective surgical procedures requiring general anesthesia (GA). Existing Intervention: The standard anesthetic regimen includes continuous intravenous infusion of the remifentanil for intraoperative analgesia, typically governed by a Target Controlled Infusion (TCI) system utilizing a pharmacokinetic/pharmacodynamic (PK/PD) model (Eleveld TCI model). Project Focus: The research seeks to improve the accuracy and efficacy of this existing analgesic strategy by integrating a multivariate patient data stream with the newly developed SEASCAPE co-pilot AI. This aims to refine the remifentanil dose predictions beyond the current TCI model's capabilities, personalized system.

Effect of Guided Imagery and AI-Assisted Video on Pain During Uterine Involution Assessment in Postpartum Women: A Randomized Controlled Trial

Uterine involution, the process by which the uterus returns to its pre-pregnancy size during the postpartum period, can cause severe cramp-like pain in postpartum women. Although pharmacological methods are common for managing this pain in the literature, their side effects limit their use in this population. This project aims to fill a significant gap in the literature by combining guided imagery, a non-pharmacological technique, with evolving artificial intelligence (AI) technologies. While the effectiveness of guided imagery in cesarean and labor pain is well-known, its impact on pain and physiological parameters during uterine involution-specifically through AI-supported visualization-will be evaluated for the first time within the scope of this study. The primary research question of the project is: "Do guided imagery and AI-supported video applications have a healing effect on the pain levels and vital signs of postpartum women during the evaluation of uterine involution?" This research is a three-arm, randomized controlled experimental study to be conducted at Istanbul Atlas University Hospital. The sample size of the study was determined using the G\*Power 3.1 program. As a result of the a priori power analysis performed by selecting the "ANOVA: Fixed effects, omnibus, one-way" statistical test under the "F tests" family, the significance level was set at α = 0.05, the statistical power (1-β) at 0.80, and the effect size (f) at 0.25 (medium effect). The analysis indicated a total sample size of 159 participants for the three groups, with 53 participants required in each group. To account for potential sample attrition, a 10% addition was made to each group, resulting in a planned study with 59 participants per group and a total of 177 participants. "Guided Imagery" and "AI-Supported Video" applications will be utilized as intervention methods. In the guided imagery group, an expert-approved original audio script titled "The Emotional Journey of Mother and Baby" will be played to the women via over-ear headphones during the uterine involution assessment. In the AI-supported video group, a video generated with AI prompts based on the same script will be shown using virtual reality (VR) goggles during the assessment. This will allow for a comparative analysis of the effects of visual and auditory stimuli on pain during the uterine involution evaluation. Participants' pain levels will be measured using the VAS scale, and physiological data (pulse, blood pressure, SpO₂) will be recorded by monitoring at the 60th, 120th, and 180th seconds of the involution process. Considering the moderating effect of anxiety on pain threshold and vital signs, the 'Postpartum Specific Anxiety Scale' will be administered to all participants before the intervention. Data will be analyzed using ANOVA and Kruskal-Wallis tests in the SPSS 23 program.

Research on Construction and Verification of Multimodal Medical Imaging Large Model

With the accumulation of multimodal clinical data such as medical imaging and electronic health records (EHRs), efficient utilization of multi-source information to achieve precise diagnosis and intelligent decision-making has become a core direction of medical artificial intelligence (AI). Although traditional unimodal algorithms have yielded outcomes in specific tasks, their inability to model the semantic correlations among imaging, textual, and laboratory data leads to insufficient stability and limited interpretability of diagnostic results, making it difficult to meet the needs of comprehensive decision-making in complex clinical scenarios. In recent years, multimodal large models have demonstrated excellent cross-modal understanding and knowledge transfer capabilities in natural images and general vision-language tasks, providing a new paradigm for medical AI. However, direct application in medical scenarios still faces challenges: first, the medical semantic system differs significantly from general language models, hindering the accurate representation of disease characteristics and imaging details; second, the complex morphology of lesions and uneven sample distribution in medical data increase the difficulty of model generalization; third, clinical data involves privacy, so data security and ethical compliance serve as prerequisites for research. The research on medical multimodal large models aims to integrate multi-source heterogeneous medical data, establish a unified semantic representation and reasoning mechanism, and realize full-process intelligent analysis including disease identification and lesion localization. This approach can not only improve the efficiency and accuracy of clinical diagnosis but also provide clinicians with interpretable and traceable auxiliary decision support, boasting broad application prospects. Based on the hospital's clinical data resources and the research team's algorithmic foundation, this study intends to construct a multimodal large model system for medical imaging diagnosis, enabling closed-loop intelligent analysis from multimodal information fusion to diagnostic report generation. The research will strictly adhere to medical ethical standards, protect patients' right to information, right to privacy, and data security. Before the official launch of the project, ethical review must be passed, and relevant regulations shall be followed to ensure the unity of scientific research and ethics, laying a compliant foundation for subsequent clinical validation and promotion.