Signals, in order.

This prospective, multinational, multicenter observational study follows patients with non-foveal, non-vision compromising atrophic age‑related macular degeneration for 24 months with 6‑month intervals. It aims to assess individual progression rates, identify focal and global retinal alterations, and evaluate monitoring using approved AI algorithms through routine imaging (Scanning Laser Fundus Photography, Color Fundus Photography, OCT, OCTA). No intervention is performed during the study, though treatment may be provided if approved in the EU.

Source: ClinicalTrials.gov

This research, conducted in the United States, examines the limited evidence supporting current clinical guidelines for diagnosing and treating rotator cuff injuries. The study compares the performance of various machine learning models that use pre‑operative data from an international, multicentric database to predict whether patients who undergo rotator cuff repair will achieve the minimal important change (MIC) for the single assessment numeric evaluation (SANE) at one‑year follow‑up.

Source: ClinicalTrials.gov

The ADOmiARN study is a multicentre, prospective, longitudinal, non-interventional observational study conducted in France and Belgium. It evaluates the in vitro diagnostic medical device EndoTest® in adolescents aged 10 to 19 years with suspected or formally diagnosed endometriosis, while keeping the usual care pathway unchanged. The study collects saliva samples and self-reported symptom and quality-of-life questionnaires without altering standard treatments or diagnostic examinations.

Source: ClinicalTrials.gov

This study evaluates if it is possible to identify quantitative parameters from audio signals to describe the changes in patient's state in relation to frailty and distress. It is conducted in the United States and falls under the research category.

Source: ClinicalTrials.gov

This research aims to develop AI-based models that predict treatment engagement and outcomes for patients in digital psychotherapy. It will use a large, real-world dataset to train the algorithms and will incorporate responsible AI practices by describing, accounting for, and mitigating bias related to mental disturbance severity and other sensitive attributes such as gender, ethnicity, and socio-demographic status.

Source: ClinicalTrials.gov

This research project in the United States aims to develop machine learning models that predict patients’ BMI and complications following bariatric surgery, forming the CABS-Score. It also seeks to predict remission rates for diabetic (DM II) patients after surgery. The resulting models will be made publicly available through a web-based application.

Source: ClinicalTrials.gov

The study evaluates an AI algorithm designed to optimize mechanical ventilation settings in ICU patients, aiming to reduce organ damage by learning from deviations between automated recommendations and expert adjustments. It highlights that mechanical ventilation, while life-saving, can be harmful if not set correctly, and that current practices often fall short of guidelines. The research proposes that AI can improve patient outcomes by tailoring ventilator and extracorporeal lung support settings to individual clinical characteristics.

Source: ClinicalTrials.gov

The study evaluates the feasibility of a new method for self-recordable ECGs using a smartwatch coupled with an image processing algorithm. The long-term goal is to establish this method and potentially integrate it into telemedical care.

Source: ClinicalTrials.gov

Idiopathic pulmonary fibrosis, non‑specific interstitial pneumonia, and chronic obstructive pulmonary disease are severe, progressive, irreversible pulmonary disorders with modest therapeutic response and poor prognosis. Prompt and accurate diagnosis is essential to enable early care that can delay disease progression and prolong survival. Artificial intelligence–assisted digital lung auscultation and deep‑learning automation of lung ultrasound offer alternative, objective methods for earlier diagnosis of these conditions.

Source: ClinicalTrials.gov

The study aims to create a predictive modeling service for postoperative complications and prolonged hospital stay in patients with psychiatric diseases undergoing colorectal surgery. It also investigates how preoperative risk factors, psychiatric and psychosomatic diseases affect colorectal surgery outcomes and complications such as anastomosis insufficiency using predictive modeling techniques. The resulting service will be publicly available as a web-based application.

Source: ClinicalTrials.gov