Clinical Research Directory

AI Ambient Scribe in Psychiatry

The goal of this clinical trial is to learn if an artificial intelligence (AI) tool that listens to appointments and drafts clinical notes can help clinicians in psychiatric care. Researchers also want to know if the tool changes how patients experience their care. The study includes mental health care providers (physicians, psychologists/psychotherapists, nurses, and social workers) at Psychiatrie St. Gallen and the adult patients they see in their appointments. The main questions it aims to answer are: * Does the AI tool lower the time clinicians spend writing notes? * Are notes written with the AI tool as good as or better than notes written without it? * Do patients notice a difference in their relationship with their clinician when the tool is used? Researchers will compare clinicians who use the AI tool to clinicians who keep writing notes their usual way, before and during the intervention period, to see if writing time, note quality, and the patient-rated therapeutic relationship differ between the two groups. Clinicians will: Join one of two groups. One group starts using the AI tool on August 14, 2026. The other group keeps writing notes the usual way. Both groups will be observed until the end of November 2026, with note quality evaluation completed in December 2026. Clinicians who cannot contribute to all measurements may, with their consent, remain in the study for automated measurement of documentation time only. Patients will: * Be invited to fill out a short, anonymous survey about their experience with their clinician. * Be asked to agree, in writing, that researchers may review the report from their initial assessment appointment.

Effects of Agent-assisted, LLM-assisted and Traditional Workflows on Diagnosis and Management Planning at Admission

The goal of this clinical trial is to evaluate whether AI-assisted workflows improve physicians' admission diagnosis and management planning performance on standardized simulated inpatient cases, among practicing internal medicine and surgery physicians across all seniority levels and across three tiers of the Chinese healthcare system. The main questions it aims to answer are: * Does the Agent-assisted workflow yield better structured admission diagnosis and management planning scores than standalone LLM assistance? * Does the Agent-assisted workflow outperform the traditional workflow without AI tools? Researchers will compare three parallel groups (traditional workflow group, LLM-assisted group, Agent-assisted group) to determine whether the Agent tool can improve diagnostic accuracy and efficiency. Participants will: * Be recruited from 15 hospitals in China and participate remotely under video proctoring * Be randomly assigned to one of the three fixed workflows, with randomization stratified by hospital tier, specialty and seniority * Complete 6 anonymized simulated HIS admission cases within one hour * Submit structured answers for each case covering principal diagnosis, secondary diagnoses, differential diagnoses, diagnostic justification, next diagnostic or therapeutic steps, consultation and referral decisions, and diagnostic confidence * Have their operation logs and time consumption recorded automatically by the study platform

Video/Image Library of Endoscopy Procedures for the Development of AI-empowered Endoscopy Quality Reporting and Educational Modules

The goal of this observational study is to establish a video/image library dataset of complete endoscopy or partial colonoscopy procedures for patients with rectal cancer or inflammatory bowel disease (IBD). With this video/image library, the aims are: * to develop and validate novel AI-empowered solutions to automatically detect and report endoscopy quality metrics * to develop automated endoscopy reporting solutions, auditing, and educational tools for residents and fellows to enhance their endoscopy skills. The hypothesis is that a heterogeneous video/image library will provide: * comprehensive and robust source material to develop AI models * real-time quality feedback at the end of an endoscopy procedure.

Artificial Intelligence in Perioperative Nursing

This mixed-methods study aims to assess current perspectives, attitudes, and preparedness of perioperative nurses regarding the integration of artificial intelligence (AI) in clinical practice. The study targets nurses working in surgical wards and operating rooms to explore AI utilization, perceived usability, professional impact, and readiness for future implementation. Quantitative and qualitative data will be collected concurrently and integrated to generate comprehensive insights into AI adoption and future directions in perioperative nursing.

Prospective User Study and Multicenter Validation of Multimodal Medical Imaging Large Models

Following model development and locking, the fixed model is evaluated in prospectively collected CT cohorts from two centers. The study is observational and does not affect clinical care. A subset of cases is used in a randomized crossover reader study.

Postpartum Education Via Artificial Intelligence for Recovery and Loneliness: A Randomized Controlled Trial

The goal of this clinical trial is to learn whether a postpartum chatbot powered by generative artificial intelligence (genAI) can help new mothers get better pelvic floor health information and feel less lonely after childbirth. The main questions this study aims to answer are: * Does using the chatbot improve postpartum pelvic floor health knowledge? * Does using the chatbot help reduce feelings of loneliness during the postpartum period? * Does using the chatbot impact pelvic floor symptoms? Researchers will compare standard postpartum care to standard care plus the chatbot. Participants will: Be assigned by chance (like flipping a coin) to standard postpartum care with or without access to the chatbot. If in the chatbot group, participants will receive education and support via the chatbot over a 4-week period. Both groups will complete questionnaires to measure their pelvic floor knowledge, pelvic floor symptoms, feelings of loneliness, depression, infant bonding, perceived social support, adverse childhood experiences, and peri-traumatic distress. The chatbot was created by urogynecology experts in collaboration with UC San Diego computer science and biomedical informatics researchers. The chatbot is designed to give new mothers personalized, evidence-based information and support in real time.

Artificial Intelligence Supported Case Analysis and Nursing Students

In this study, it was aimed to evaluate the effect of artificial intelligence-supported case analysis method on nursing students' knowledge, case management performances and nursing diagnosis determination skills.This study was conducted in a single-blind randomised controlled trial design. Students were randomly assigned to the traditional teaching group, the case analysis group and the artificial intelligence group.

AI-Supported Case Analysis Among Nursing Students

The aim of this study is to determine the effect of AI-supported internal medicine nursing case analysis on students' case management performance, learning outcomes, learning experience, clinical self-efficacy, and cognitive load levels. This study will be conducted using a single-blind randomized controlled trial design for the quantitative research and an individual interview design for the qualitative research. Students will be randomly assigned to either the intervention (artificial intelligence) or control (case analysis) group.

Development of a Prediction Score for the Occurrence of Death or Lung Transplantation in Patients With Emphysema Secondary to Alpha-1-anti-tripsin Deficiency

Emphysema linked to alpha-1-antitrypsin deficiency (DAAT): towards a better prediction of risks Emphysema caused by alpha-1-antitrypsin deficiency (DAAT) is a rare genetic disorder that can lead to serious complications, such as the need for a lung transplant or death, affecting up to 15% of patients. The only specific treatment available is a weekly infusion of alpha-1-antitrypsin (IV-AAT), an expensive and burdensome therapy. Currently, there is no reliable model to predict the course of the disease in these patients. Our study, conducted in several French hospitals, aims to develop a prediction tool combining clinical, biological, functional data and advanced medical image analysis (lung CT). This model will make it possible to identify the most at-risk patients, in order to better adapt their care, anticipate transplant needs and avoid unnecessary treatments for low-risk patients. Ultimately, this approach could also improve access to care for patients who need it most, while optimizing health system resources.

Development and Validation of an AI Foundation Model for Frozen-Section Pathology

This multicenter observational study aims to develop and validate an artificial intelligence foundation model for frozen-section pathology. The study includes a retrospective phase and a prospective validation phase. Retrospective frozen-section pathology data will be used for model development, internal validation, and external validation. A prospective multicenter cohort of patients undergoing intraoperative frozen-section examination will then be enrolled to evaluate the model in a real-world clinical setting. The model will analyze digitized frozen-section whole-slide images and will be evaluated for prespecified frozen-section pathology diagnostic tasks across multiple organ systems. Its performance will be assessed using pathological reference standards. The primary outcome is the area under the receiver operating characteristic curve. Secondary outcomes include accuracy, sensitivity, specificity, positive predictive value, and negative predictive value. This study is observational and will not require research-mandated changes to routine clinical care.

Narrow Band Imaging Versus Artificial Intelligence for Colonic Surveillance in Inflammatory Bowel Disease

Patients with inflammatory bowel disease (IBD) - ulcerative colitis or Crohn's disease - have a higher risk of developing colorectal cancer than the general population. For this reason, regular colonoscopies are recommended to look for early warning signs, such as abnormal areas of tissue called dysplasia, which can develop into cancer over time. Finding these abnormal areas during colonoscopy can be difficult because IBD causes ongoing inflammation that can make the bowel mucosa look irregular, hiding subtle changes. Doctors currently use a technique called narrow-band imaging (NBI), a special light setting on the colonoscope that enhances the visibility of the bowel, to help spot these areas more easily. A newer tool, artificial intelligence (AI)-assisted detection, has shown promise in helping doctors find more polyps during routine colonoscopies in the general population. However, this AI tool was developed and tested mostly in people without IBD, so it is not yet known whether it works as well in people with IBD, whose bowel can look very different due to chronic inflammation. This study will directly compare the AI tool (CADe; ENDO-AID, Olympus) with narrow-band imaging to see which one is better at finding abnormal areas during colonoscopy in patients with long-standing ulcerative colitis or Crohn's disease who are having their routine cancer surveillance exam. Each participant will have one colonoscopy in which the bowel is examined twice in a row, once with each technique, by two different doctors, in random order. This lets researchers compare both methods directly within the same patient, which is the fairest comparison. The study aims to enroll 60 patients at Vall d'Hebron University Hospital in Barcelona, Spain. Researchers hope the results will help determine whether AI tools - which are widely available and easier to use than NBI - can be a reliable alternative for IBD surveillance, potentially making this important cancer-screening exam more accessible in the future.

AI Virtual Patient Training for Dental Student History Taking

This randomized study evaluated whether an artificial intelligence-assisted virtual patient could support third-year dental students in practicing medical and dental history taking. Fifty-six students were assigned to either a Gemini-based virtual-patient training group or a conventional role-play group. Both groups completed four training sessions over 2 weeks using comparable clinical cases, practice time, and feedback criteria. The study compared changes in case-based history-taking competency as well as student satisfaction and perceived usefulness of the training methods.

AI Timing in Chest X-ray Interpretation Using Eye-Tracking

Chest X-rays are commonly used to help diagnose and manage chest conditions. Artificial intelligence (AI) tools are increasingly being used to support chest X-ray interpretation. However, it is not yet clear whether the timing of AI information affects how clinicians review images, make decisions, and use AI support. This study will look at whether showing AI information before or after a clinician first reviews a chest X-ray changes how they look at the image, how long they take, their interpretation decisions, their confidence, and their trust in AI support. Healthcare professional participants will complete two chest X-ray interpretation sessions in a controlled NHS research setting. During each session, participants will review de-identified chest X-ray images while wearing eye-tracking equipment. Eye-tracking will record where a participant looks on the image and how long they spend looking at different areas. In one session, AI information will be shown before the participant reviews the chest X-ray. In the other session, AI information will be shown after the participant has first reviewed the chest X-ray. The order of these two sessions will be balanced across participants. The study uses de-identified chest X-ray images from existing examinations. It does not involve patients directly, does not change clinical care, and no clinical decisions will be made from the study readings. Participants will also complete a short questionnaire about their experience of using AI support. A separate anonymous survey will collect wider views from clinicians, patients, members of the public, and healthcare staff about the use of AI in chest X-ray interpretation.

The Effect of Artificial-Intelligence-Assisted Videos on Breastfeeding Self-Efficacy, Motivation, and LATCH Scores in First-Time Mothers

As of 2024, nearly half (48%) of infants under six months worldwide are exclusively breastfed, approaching the global target of 50%. Building on this progress, the World Health Organization has extended the target to 60% by 2030, emphasizing the need for innovative, scalable, and supportive interventions to strengthen breastfeeding practices. Breastfeeding has well-established benefits for infant growth, immunity, and long-term health, while also reducing maternal postpartum complications and chronic disease risks. Early postpartum support, particularly within the first hours after birth, is critical for successful and sustained breastfeeding. However, in busy clinical settings, providing continuous and individualized support can be challenging, especially for primiparous women who may experience low confidence, pain, and insufficient guidance. This randomized controlled trial aims to evaluate the effect of an artificial intelligence (AI)-supported relaxing breastfeeding video on breastfeeding self-efficacy, breastfeeding motivation, and LATCH scores among primiparous women. Unlike instructional videos, the AI-based video is designed to promote emotional relaxation, instinctive breastfeeding perception, and maternal confidence during the early postpartum period. The study adopts a two-arm randomized controlled experimental design. The population consists of primiparous women who deliver vaginally at Ağrı Training and Research Hospital postpartum unit between February and June 2026. A priori power analysis (α=0.05, power=0.95) indicated a minimum sample size of 38 participants; considering a 20% attrition rate, a total of 46 women (23 per group) will be recruited. Eligible participants include primiparous, Turkish-speaking women without postpartum or neonatal complications. Women who undergo cesarean delivery, have medical or psychiatric conditions preventing breastfeeding, or whose newborns require intensive care will be excluded. Participants will be randomized into intervention and control groups using an online randomization tool. All participants will receive a standardized 5-minute breastfeeding education based on the Turkish Ministry of Health breastfeeding counseling guidelines. In addition to standard care, the intervention group will watch a 10-minute AI-supported relaxing video at the 2nd and 6th postpartum hours during breastfeeding. The video will be displayed via tablet while the mother is in a comfortable breastfeeding position. The control group will receive standard care only. The AI-generated video will be produced using Kling AI, a generative video platform that enables controlled text-to-video workflows. To ensure ethical and cultural sensitivity, the video will not include real human or animal breastfeeding images. Instead, it will feature abstract, metaphorical visuals (e.g., pastel silhouettes, minimalist line art, or flat illustrations) that convey calmness, bonding, rhythm, and instinctive closeness. The final version will be selected following expert review and pilot testing with three postpartum women. Low-level white noise (\<60 dB) will accompany the video to enhance maternal relaxation and infant comfort. Data collection tools include a demographic information form, the Breastfeeding Self-Efficacy Scale-Short Form, the Primipara Breastfeeding Motivation Scale, and the LATCH Breastfeeding Assessment Tool. Breastfeeding observations and LATCH scoring will be conducted by an independent midwife blinded to group allocation. Statistical analyses will include descriptive statistics, paired and between-group comparisons, and repeated-measures analyses where appropriate. Ethical approval will be obtained from the relevant institutional ethics committee, and written informed consent will be secured from all participants. The findings are expected to contribute novel evidence on the role of AI-supported emotional and relaxing digital interventions in enhancing early postpartum breastfeeding outcomes and maternal confidence.

AI Role-Play Coaching for Breastfeeding Counseling

Breastfeeding counseling is critically important for both maintaining the mother's motivation to breastfeed and ensuring the baby's healthy nutrition. Artificial intelligence (AI)-based role-playing coaching is an innovative teaching approach that allows students to experience the counseling process through digital scenarios, identify and correct their mistakes in a risk-free environment, and receive instant feedback based on their performance. In this context, this study aims to examine the effect of AI-assisted role-playing coaching on breastfeeding counseling skills in nursing students using a randomized controlled experimental design. The study was conducted in the Nursing Laboratory of a university's Nursing Department between March 2026 and July 2026. Participants were divided into three groups: Peer role-playing group, Classical theoretical training (control) group, and AI-assisted role-playing group. Data were collected using a personal information form, the Self-Efficacy Scale for Supporting Breastfeeding Mothers, and the Breastfeeding Counseling Role-Playing Assessment Rubric. The findings show that the peer role-playing group has a higher ability to establish open, supportive, and empathetic communication with the mother, while the AI-assisted role-playing group has a higher ability to collect, evaluate, and provide feedback on data. Furthermore, breastfeeding self-efficacy was found to be higher in the AI-assisted role-playing group. The data obtained have the potential to generate evidence for the use of next-generation learning technologies in nursing education and contribute to training graduate nurses ready for clinical practice. It will also provide a scientific basis for developing training strategies aimed at improving the quality of breastfeeding support services.

Morphology in Oral Rare Syndromes & Artificial Intelligence for Clinical Diagnosis

MOSAIC aims to determine whether oro-dental morphological anomalies, particularly palatal morphology, associated with rare bone and cartilage diseases can be precisely characterized using 3D digital models analysed through geometric morphometrics. The study will also evaluate whether these morphological signatures can train an artificial intelligence (AI) algorithm to classify syndromes. A prospective monocentric case-control cohort will be constituted, including 3D intra-oral scans and associated clinical data. The final goal is to improve diagnostic accuracy and reduce diagnostic delay in rare bone disorders.

Comparison of Digital Analysis and Artificial Intelligence for Cephalometric Tracing

This study aims to evaluate the accuracy and reliability of artificial intelligence (AI)-based cephalometric analysis compared with digital manual tracing. A total of 100 standardized lateral cephalometric radiographs will be analyzed using Delta-Dent software with manual landmark identification and three fully automated AI-based systems (WebCeph, QuantX, and Smartee). Sagittal, vertical, dental, and soft tissue cephalometric parameters will be compared among the different methods. Statistical analysis will assess inter-method agreement and the clinical relevance of any observed discrepancies. The study seeks to determine whether AI-based systems provide measurements comparable to conventional digital tracing and whether they can be considered reliable adjunctive tools in orthodontic diagnosis and treatment planning.

Diagnostic Accuracy of GPT-4o and Claude for HEART Score Calculation in Chest Pain

This prospective observational diagnostic accuracy study evaluates whether large language models (LLMs) - GPT-4o (OpenAI, gpt-4o-2024-11-20) and Claude (Anthropic, claude-sonnet-4-6) - can accurately calculate HEART scores from unstructured Turkish clinical notes and predict 30-day major adverse cardiac events (MACE) in emergency department patients presenting with non-traumatic chest pain. The study will enroll 600 consecutive adult patients. For each patient, the same anonymized data (free-text anamnesis, ECG report text, troponin value, and age) will be independently processed by both LLMs via separate API calls with deterministic settings (temperature=0, JSON format). A three-expert consensus HEART score - derived through blinded independent scoring by three emergency medicine physicians with majority-vote adjudication - serves as the reference standard for agreement analysis. Actual 30-day MACE (all-cause death, AMI Type 1/2/4b, unplanned revascularization) determined via national health database and telephone follow-up serves as the outcome for diagnostic accuracy analysis. A secondary documentation-quality sub-study will quantify how spontaneously Turkish emergency anamnesis notes capture HEART score parameters.

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

This study is conducted under the ethics-approved project titled "Artificial Intelligence Solution for Simplifying the Diagnostic Workflow of Breast MRI''.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.

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.

Development and Pre-validation of a Machine Learning-based Prediction Algorithm for Early Functional Recovery in Patients Undergoing Hip and Knee Replacement Surgery

The goal of this observational study is to develop and pre-validate a machine learning algorithm to predict early recovery of mobility in patients undergoing hip or knee joint replacement surgery. The primary research question is: Can a machine learning model accurately classify patients with faster versus slower recovery of autonomous mobility in the first days after joint replacement surgery? Patients who have undergone elective hip or knee arthroplasty and received post-operative physiotherapy will have their clinical and perioperative data collected retrospectively (2020-2023) and prospectively (March 2026-December 2027). The algorithm will be trained on retrospective data and tested prospectively to evaluate its predictive performance for early mobilization and length of hospital stay.

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).

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.

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.