Citations

Abstract Objective Despite exponential growth in artificial intelligence (AI) research for laboratory medicine and pathology, a significant gap exists between model development and clinical AI implementation. This article introduces a structured framework, the Clinical AI Readiness Evaluator (CARE), to bridge this gap and support the responsible adoption of AI in clinical laboratory settings. Methods Building upon the Machine Learning Technology Readiness Levels framework, we developed CARE specifically for the clinical laboratory environment by incorporating health care–specific requirements, regulatory considerations, and workflow integration needs. This framework was iteratively refined through practical application across diverse AI use cases within laboratory medicine and pathology. Results The CARE framework provides a systematic approach to AI development and implementation through 8 component workstreams: clinical use case, data, data pipeline, code, clinical user experience, clinical technology infrastructure, clinical orchestration, and regulatory compliance. Unlike generic AI frameworks, CARE distinguishes itself by emphasizing both health care and laboratory workflow integration, regulatory requirements, ethical considerations, and comprehensive validation for clinical contexts. The framework accommodates both internally developed models and commercial AI solutions, providing clear guidance through technology readiness levels and structured review processes. Conclusions The CARE framework addresses the unique challenges of implementing AI in laboratory medicine and pathology by providing a comprehensive roadmap from initial concepts through clinical deployment and maintenance. This article, the first in a series of 4, establishes the foundational AI lifecycle framework, while subsequent articles will explore data documentation, ethical AI considerations, and governance structures. By adopting this structured approach, laboratories can responsibly harness AI’s potential to enhance diagnostic accuracy and operational efficiencies and, ultimately, improve patient care.

3069 Background: Molecular targeted agents have revolutionized cancer diagnosis and treatment. Intraoperative molecular imaging (IMI) is a novel technique that entails performing real-time, in vivo optical imaging during surgery to ensure cancer clearance. A multicenter clinical trial was conducted using abenacianine, a cathepsin targeted near-infrared (NIR) IMI agent to visualize pulmonary nodules during surgery. The primary objective was to assess the efficacy of abenacianine in localizing pulmonary lesions, evaluating surgical margin, and identifying unsuspected disease. Methods: Participants scheduled to undergo surgery for known or suspected cancer in the lung received abenacianine (0.32 mg/kg intravenous) preoperatively 12 to 36 hours before surgery. During a standard-of-care surgery, the lung underwent IMI and additional disease not identified by conventional methods were identified and analyzed. Efficacy was measured by frequency of clinically significant events (CSEs) defined as localization of lesions not found by standard surgical techniques, identification of additional cancers, identification of positive surgical margins confirmed by histology, and detection of cancerous lymph nodes. Results: 89 participants were included in the study. The mean age was 67 years and 57 (64%) were female. Of 89 participants administered abenacianine who underwent standard of care surgical resection for known or suspected cancer in the lung, 40 (45%) had at least one CSE. Abenacianine with NIR imaging identified lesions that were not found by standard surgical methods in 34 (38%) participants, synchronous and occult cancers that were not found by pre-operative imaging in 2 (2%), margins within 10 mm of the closest staple line in 8 (9%), and lymph nodes determined to be cancerous in 1 (1%) participant. Tumors visualized by IMI with abenacianine included non-small cell lung cancers (adenocarcinoma, squamous cell carcinoma, neuroendocrine tumor) and cancers that metastasized to the lung (breast, colorectal, prostate, thymoma, renal cell, sarcoma). Abenacianine was safe and well tolerated in this study; there were no drug-related serious adverse events. Conclusions: This Phase 2 multicenter study demonstrated that using IMI with abenacianine during standard-of-care lung cancer surgery markedly improved clinical outcomes. Abenacianine localized tumors intraoperatively, identified synchronous and occult lesions, helped assess negative margin status, and identified cancerous lymph nodes enabling a more complete oncologic resection. Clinical trial information: NCT06145048 .

Abstract Purpose: Soft-tissue sarcomas are rare malignancies with poor prognosis and limited systemic treatment options. We conducted a phase II study to assess the efficacy and safety of trabectedin and olaparib in patients with advanced disease. Patients and Methods: Patients with soft-tissue sarcoma who received ≥1 prior therapy were recruited into two cohorts. Cohort 1 included leiomyosarcoma and liposarcoma; cohort 2 included other histologies. All patients received trabectedin (1.1 mg/m2 24-hour infusion every 21 days) and olaparib (150 mg twice daily continuously). The study was conducted using a Simon minimax two-stage design with a primary endpoint of objective response (OR) rate per RECIST 1.1. Results: Twenty-nine (cohort 1, 16; cohort 2, 13) patients enrolled; one patient in cohort 2 was not evaluable. There were no confirmed ORs in cohort 1; the best response was stable disease in 12 (75%) patients and progressive disease in four (25%). Two partial responses were observed in cohort 2 (n = 12). The most common adverse events were fatigue (75%), neutropenia (75%), anemia (68%), and thrombocytopenia (68%). The median progression-free survival and overall survival for all patients were 3.5 (95% confidence interval, 3.3–8.2) and 13.2 months (95% confidence interval, 10.3–20.9), respectively. Next-generation sequencing of 17 tumors revealed multiple abnormalities, most commonly in TP53, RB1, and ATRX. Conclusions: Trabectedin plus olaparib conferred high rates of toxicity and failed to demonstrate ORs in leiomyosarcoma and liposarcoma. Preliminary evidence of clinical benefit in two patients in cohort 2 suggests potential value of either or both drugs in other sarcomas.

Abstract An individual participant data (IPD) meta-analysis can assess the predictive value of data on outcomes at the individual level, offering a potential tool for developing personalized pain management. Pretreatment quantitative sensory testing (QST) may stratify patient groups, which are then linked to treatment outcomes. Our objective was to determine if measures of QST at baseline are related to treatment outcomes (at any time point) for pain and disability in lower-limb osteoarthritis. We performed a systematic review with an IPD meta-analysis. Searches were conducted in 9 databases until May 5, 2023 for intervention studies that measured baseline QST and longitudinal measures of participant-reported pain and disability. We performed a 2-stage approach to analyse longitudinal data. Individual models were fitted to each study and combined using random effects multivariate meta-analytic models. Study quality was assessed using the Joanna Briggs Institute checklist, and certainty of the evidence was assessed using GRADE. We identified 3082 records and included 1 hip and 28 knee datasets consisting of 2522 participants from 40 studies. Local warm detection thresholds (P = 0.024) predicted knee osteoarthritis pain outcomes (very-low certainty). Local warm detection thresholds (P = 0.030), remote cold detection thresholds (P = 0.05), and remote pressure tolerance thresholds (P = 0.007) predicted knee osteoarthritis disability outcomes (very-low certainty). Other QST variables were associated with hip and knee osteoarthritis pain and disability levels (eg, pressure pain thresholds), but this relationship did not change over time. This review finds that mechanism-based, QST methodologies do not consistently predict pain or disability on an individual level in hip or knee osteoarthritis.

AbstractIntroductionActive learning engages learners in constructing knowledge through interactive strategies such as simulation, small‐group discussion, and peer instruction. Although recognized as a superior approach to traditional passive learning, its adoption has been inconsistent. Barriers include reliance on traditional lectures, lack of training, and limited time to develop materials. Simulation educators have a unique skill set that may inform and support active learning initiatives.MethodsFifteen emergency medicine simulation experts convened to define the unique skill set of simulation educators and identify transferable simulation‐based medical education (SBME) skills and concepts to promote effective active learning beyond the simulation lab. Workgroup members are simulation education directors who have undergone specialized training in SBME. During biweekly meetings over 6 months, workgroup members reviewed primary literature in SBME and active learning, along with relevant simulation educator training materials. Objectives were achieved through iterative review, group conceptualization, and expert consensus.Unique treatmentThe increasing prevalence of SBME in medical education, along with the growth of simulation fellowships, has produced a group of education experts with shared competencies. Simulation educators acquire expertise in psychological safety, facilitation, communication, and debriefing, through specialized training and extensive experience. These skills are critical for active learning environments where learners benefit from structured, engaging, and psychologically safe experiences.Implications for educatorsKey transferable SBME topics were identified for use in active learning environments outside of the simulation lab: (1) psychological safety, (2) facilitation strategies, and (3) communication techniques. Transferable tools and concepts were identified to promote efficacy and learner engagement during active learning in diverse environments. Simulation educators’ expertise is an underutilized resource for faculty development initiatives aimed at advancing active learning. This work advocates for leveraging simulation educators’ skills to close the active learning implementation gap, enhance learner outcomes, and meet the evolving needs of contemporary medical education.

Inadequate light penetration in tissues restricts photodynamic therapy to treating only superficial tumors. To enable x-ray–excited photodynamic therapy (XPDT) that targets deep-seated tumors, we synthesized a nanoscintillator-photosensitizer complex containing 5% Eu-doped Y 2 O 3 fluorescing at 611 nanometers and decorated with SiO 2 containing the scintillation-coupled photosensitizer methylene blue and a polyethylene glycol coating [PEGylated Y 2 O 3 :Eu@SiO 2 -methylene blue (pYSM)]. When irradiated, pYSMs generate singlet oxygen species in vitro, causing cytotoxicity with hallmarks of immunogenic cell death (calreticulin translocation to the cell membrane). Intravenously administered pYSMs home passively to pancreatic tumor xenografts and, upon 10 gray irradiation, cause significant tumor regression ( P < 0.01). On combining XPDT with anti-PD1 immunotherapy, a distant nonirradiated tumor also regresses via an increase in intratumoral activated CD8 + cytotoxic T cells. Collectively, we advance a systemically delivered XPDT strategy that mediates an antitumor effect in both irradiated and nonirradiated (abscopal) tumors when coupled with immunotherapy, converting an immunologically “cold” tumor to an immunologically “hot” tumor.

ABSTRACT Background Mitral annular disjunction (MAD) carries an increased risk of complex ventricular arrhythmias, which can lead to sudden cardiac death. Many of these patients undergo implantable cardioverter defibrillator (ICD) implantation, but their ICD outcomes are not known. Objective The aim of this study was to assess the outcomes of ICD implantation and the predictors of appropriate ICD therapies in patients with MAD. Methods The study included patients with MAD who underwent ICD implantation. Clinical, electrocardiographic, cardiac imaging, and device therapy data were collected. Results Forty‐nine patients with MAD and ICD were included. Median age was 49 (21) years, and 29 (59%) were female. 13 (27%) patients underwent ICD implantation for primary prevention and 36 (73%) patients for secondary prevention. Over a median follow‐up of 27.3 (35.3) months, 23 (47%) patients received ICD therapies. 18 (37%) patients had appropriate ICD therapies, and 5 (10%) patients had inappropriate ICD shocks. Median time to first appropriate therapy was 22.3 (63.3) months. In patients with a secondary prevention ICD indication, the rate of appropriate ICD therapies was 44%, while in patients with a primary prevention ICD indication, it was 15%. Among patients with appropriate ICD therapies, the first therapies were delivered for monomorphic ventricular tachycardia (VT) in 7 (39%) patients and polymorphic VT or ventricular fibrillation (VF) in 11 (61%) patients. Patients with appropriate ICD therapies were more likely to have a history of SCA ( p = 0.003) and/or low left ventricular ejection fraction (LVEF) ( p = 0.022) before ICD implantation as compared to patients without appropriate ICD therapies. Conclusions In our cohort of patients with MAD and ICD, appropriate ICD therapies were common. Most appropriate ICD therapies were delivered for polymorphic VT or VF. Larger studies are needed to elucidate the mechanisms of VAs and refine risk stratification in MAD.

OBJECTIVES: This study aimed to establish a set of guiding principles for data sharing and harmonization in critical care, focusing on the use of real-world data (RWD) and real-world evidence (RWE) to improve patient outcomes and research efficacy. The principles were developed through a systematic literature review and a modified Delphi process, with the goal of enhancing data accessibility, standardization, and interoperability across critical care settings. DATA SOURCES: Data sources included a comprehensive search of peer-reviewed literature, specifically studies related to the use of RWD and RWE in healthcare, guidelines, best practices, and recommendations on data sharing and harmonization. A total of 8150 articles were initially identified through databases such as MEDLINE and Web of Science, with 257 studies meeting inclusion criteria. STUDY SELECTION: Inclusion criteria focused on publications discussing health-related informatics, recommendations for RWD/RWE usage, data sharing, and harmonization principles. Exclusion criteria ruled out non-human studies, case studies, conference abstracts, and articles published before 2013, as well as those not available in English. DATA EXTRACTION: From the 257 selected studies, 322 statements were extracted. After removing irrelevant definitions and off-topic content, 232 statements underwent content validation and thematic analysis. These statements were then consolidated into 24 candidate guiding principles after rigorous review and consensus-building among the expert panel. DATA SYNTHESIS: A three-phase modified Delphi process was employed, involving a conceptualization group, a review group, and a Delphi group. In phase 1, experts identified key themes and search terms for the systematic review. Phase 2 involved validating and refining the prospective guiding principles, while phase 3 employed a Delphi panel to rate principles on acceptability, importance, and feasibility. This process resulted in 24 guiding principles, with high consensus achieved in rounds 2 and 3 on their relevance and applicability. CONCLUSIONS: The systematic review and Delphi process resulted in 24 guiding principles to improve data sharing and harmonization in critical care. These principles address challenges across the data lifecycle, including generation, storage, access, and usage of RWD and RWE. This framework is designed to promote more effective and equitable data practices, with relevance for the development of artificial intelligence-based decision support tools and clinical research. The principles are intended to guide the responsible use of data science in critical care, with emphasis on ethics and equity, while acknowledging the variability of resources across settings.

OBJECTIVES: To develop the first core Critical Care Data Dictionary (C2D2) with common data elements (CDEs) to characterize critical illness and injuries. DESIGN: Group consensus process using modified Delphi approach. SETTING: Electronic surveys and in-person meetings. SUBJECTS: A multidisciplinary workgroup of clinicians and researchers with expertise in the care of the critically ill and injured. INTERVENTIONS: The Delphi process was divided into domain and CDE portions with each composed of two item generation rounds and one item reduction/refinement rounds. Two in-person meetings augmented this process to facilitate review and consideration of the domains and by panel members. The final set of domains and CDEs was then reviewed by the group to meet the competing criteria of utility and feasibility, resulting in the core dataset. MEASUREMENTS AND MAIN RESULTS: The 23-member Delphi panel was provided 1833 candidate variables for potential dataset inclusion. The final dataset includes 226 patient-level CDCs in nine domains, which include anthropometrics and demographics (8), chronic comorbid illnesses (18), advanced directives (1), ICU diagnoses (61), diagnostic tests (42), interventions (27), medications (38), objective assessments (26), and hospital course and outcomes (5). Upon final review, 91% of the panel endorsed the CDCs as meeting criteria for a minimum viable data dictionary. Data elements cross the lifespan of neonate through adult patients. CONCLUSIONS: The resulting C2D2 provides a foundation to facilitate rapid collection, analyses, and dissemination of information necessary for research, quality improvement, and clinical practice to optimize critical care outcomes. Further work is needed to validate the effectiveness of the dataset in a variety of critical care settings.

BACKGROUND: Patients with arrhythmogenic mitral valve prolapse syndrome are at increased risk for life-threatening ventricular arrhythmias, but studies have been limited by small sample sizes. We sought to assemble an international arrhythmogenic mitral valve prolapse syndrome registry to delineate the clinical, imaging, and treatment characteristics of patients with arrhythmogenic mitral valve prolapse syndrome who survived sudden cardiac arrest (SCA) or had sustained ventricular tachycardia (VT) or ventricular fibrillation. METHODS: In this descriptive registry, we characterized patients with arrhythmogenic mitral valve prolapse syndrome who survived SCA, sustained VT, or ventricular fibrillation. Deidentified data were abstracted locally and combined centrally. RESULTS: We included 148 patients who had SCA or VT/ventricular fibrillation. Patients had a mean age of 43.7±15.4 years; 68% were women, 73% had bileaflet prolapse, 65% had mitral annular disjunction, 67% had nonsustained VT, and 59% had inferolateral T-wave inversions. Syncope (n=54, 48%) and anterolateral T-wave inversion (n=26, 22%) were relatively common. Catheter ablation was performed in 50 (35%) patients for premature ventricular complexes and in 18 (17.7%) patients for VT. Sites of origin for arrhythmias were commonly in the papillary muscles, fascicles, mitral annulus, and inferior/inferolateral left ventricle. CONCLUSIONS: In this international descriptive registry of patients with arrhythmogenic mitral valve prolapse syndrome and SCA, patients were young, women, and had bileaflet mitral valve prolapse, mitral annular disjunction, inferolateral T-wave inversions, and nonsustained VT. A history of syncope and anterolateral T-wave inversions was relatively common in patients who survived SCA or sustained VT/ventricular fibrillation.