Try a new search

Format these results:

Searched for:

person:oermae01

in-biosketch:true

Total Results:

170


LLM ensemble PREVIA predicts future PTSD risk from emergency department trauma narratives with expert-level accuracy

Mueller, Victoria; Yadav, Jyoti; Farnsworth, Annie; Hilberdink, Charlotte E; Gershov, Sapir; Oermann, Eric K; Wall, Stephen P; Shalev, Arieh Y; Schultebraucks, Katharina
Post-traumatic stress disorder affects numerous trauma survivors following emergency department discharge, making this encounter vital for early risk stratification. Yet emergency clinicians face severe time constraints, making existing screening tools impractical. We introduce PREVIA, a clinically-informed large language model ensemble, that predicts post-traumatic stress disorder risk from unstructured patient trauma narratives, collected in response to a routine-care question asked in the emergency department, by extracting early post-traumatic stress symptoms, peri-traumatic symptoms, risk, and protective factors. Within an ongoing longitudinal study, we included N = 145 patients across two New York City level-1 trauma centers within 24 hours of trauma and assessed post-traumatic stress disorder 1-month post-trauma. In this work, PREVIA accurately forecasts post-traumatic stress disorder (AUROC = 78.49% ± 5.75%, sensitivity=95.4% ± 3.99%, precision=46.76% ± 8.66%), performs comparably to an emergency clinician specialized in post-traumatic stress disorder on 30 randomly selected patients (PREVIA: AUROC = 84.39%, sensitivity=100.00%, precision=50.00%; clinician: AUROC = 83.86%, sensitivity=88.89%, precision=50.00%), and generates interpretable explanations without additional clinical burden. These initial findings support PREVIA's potential as a scalable emergency department screener for timely, targeted interventions, and warrant validation in larger, independent cohorts.
PMCID:13748573
PMID: 43003853
ISSN: 2041-1723
CID: 6075768

LLM ensemble PREVIA predicts future PTSD risk from emergency department trauma narratives with expert-level accuracy

Mueller, Victoria; Yadav, Jyoti; Farnsworth, Annie; Hilberdink, Charlotte E; Gershov, Sapir; Oermann, Eric K; Wall, Stephen P; Shalev, Arieh Y; Schultebraucks, Katharina
Post-traumatic stress disorder affects numerous trauma survivors following emergency department discharge, making this encounter vital for early risk stratification. Yet emergency clinicians face severe time constraints, making existing screening tools impractical. We introduce PREVIA, a clinically-informed large language model ensemble, that predicts post-traumatic stress disorder risk from unstructured patient trauma narratives, collected in response to a routine-care question asked in the emergency department, by extracting early post-traumatic stress symptoms, peri-traumatic symptoms, risk, and protective factors. Within an ongoing longitudinal study, we included N = 145 patients across two New York City level-1 trauma centers within 24 hours of trauma and assessed post-traumatic stress disorder 1-month post-trauma. In this work, PREVIA accurately forecasts post-traumatic stress disorder (AUROC = 78.49% ± 5.75%, sensitivity=95.4% ± 3.99%, precision=46.76% ± 8.66%), performs comparably to an emergency clinician specialized in post-traumatic stress disorder on 30 randomly selected patients (PREVIA: AUROC = 84.39%, sensitivity=100.00%, precision=50.00%; clinician: AUROC = 83.86%, sensitivity=88.89%, precision=50.00%), and generates interpretable explanations without additional clinical burden. These initial findings support PREVIA's potential as a scalable emergency department screener for timely, targeted interventions, and warrant validation in larger, independent cohorts.
PMCID:13748573
PMID: 43003853
ISSN: 2041-1723
CID: 6075770

LLM ensemble PREVIA predicts future PTSD risk from emergency department trauma narratives with expert-level accuracy

Mueller, Victoria; Yadav, Jyoti; Farnsworth, Annie; Hilberdink, Charlotte E; Gershov, Sapir; Oermann, Eric K; Wall, Stephen P; Shalev, Arieh Y; Schultebraucks, Katharina
Post-traumatic stress disorder affects numerous trauma survivors following emergency department discharge, making this encounter vital for early risk stratification. Yet emergency clinicians face severe time constraints, making existing screening tools impractical. We introduce PREVIA, a clinically-informed large language model ensemble, that predicts post-traumatic stress disorder risk from unstructured patient trauma narratives, collected in response to a routine-care question asked in the emergency department, by extracting early post-traumatic stress symptoms, peri-traumatic symptoms, risk, and protective factors. Within an ongoing longitudinal study, we included N = 145 patients across two New York City level-1 trauma centers within 24 hours of trauma and assessed post-traumatic stress disorder 1-month post-trauma. In this work, PREVIA accurately forecasts post-traumatic stress disorder (AUROC = 78.49% ± 5.75%, sensitivity=95.4% ± 3.99%, precision=46.76% ± 8.66%), performs comparably to an emergency clinician specialized in post-traumatic stress disorder on 30 randomly selected patients (PREVIA: AUROC = 84.39%, sensitivity=100.00%, precision=50.00%; clinician: AUROC = 83.86%, sensitivity=88.89%, precision=50.00%), and generates interpretable explanations without additional clinical burden. These initial findings support PREVIA's potential as a scalable emergency department screener for timely, targeted interventions, and warrant validation in larger, independent cohorts.
PMCID:13748573
PMID: 43003853
ISSN: 2041-1723
CID: 6075769

A global digital navigator of human health for precision medicine

Zhang, Kang; Dong, Jiahong; Wong, Hang; Liu, Sian; Wang, Kai; Wu, Wei; Zhou, Kaixin; Xue, Kanmin; Zhang, Edward; Gootenberg, Jonathan; Abudayyeh, Omar O; Williams, Michelle; Shen, Xian; Loupy, Alexandre; Ideker, Trey; Tan, Patrick; Wu, Joseph; Sodhi, Akrit; Oermann, Eric; Xu, Tao; Ning, Guang
PMID: 42728381
ISSN: 1546-170x
CID: 6072283

Reply to: Limited benchmarks constrain the conclusions of a general-purpose versus clinical AI comparison [Letter]

Vishwanath, Krithik; Aphinyanaphongs, Yindalon; Oermann, Eric Karl
PMID: 42693298
ISSN: 1546-170x
CID: 6072018

Advancing cancer detection and treatment using longitudinal routine clinical data

Liu, Fei; Wang, Kai; Xu, Hui; Tang, Cheng; Shen, Xian; Wang, Meihao; Yang, Lei; Yang, Li; Liu, Li; Hu, Changxi; Li, Gen; Wu, Wei; Zou, Zixing; Li, Bingzhou; Liu, Sian; Kang, Jin; Kong, Jungho; Li, Ting; Wong, Io Nam; Huang, Xiaoying; Chen, Gang; Lu, Wenyang; Ziyar, Ian; Zhang, Charlotte L; Sun, Yiwen; Lin, Weihong; Ou, Caiwen; Fok, Manson; Hou, Taiwa; Wang, Winston; Xue, Kanmin; Yin, Yun; Zhu, Hao; Gootenberg, Jonathan; Abudayyeh, Omar O; Karin, Michael; Loupy, Alexandre; Rasko, John E J; Ideker, Trey; Luo, Huiyan; Oermann, Eric; Zhang, Kang; ,
Cancer management remains fragmented across its continuum, from late-stage diagnosis and salvage therapies to non-personalized surveillance. Here, we present Oncoformer, a unified multimodal transformer model trained on the China Oncology Multimodal Prediction and Surveillance Study (COMPASS) cohort (3.67 million individuals, 17.7 million clinical visits) and validated on independent external cohorts, including the UK Biobank. Oncoformer integrates longitudinal electronic health records with chest X-ray imaging to address multiple clinical tasks: pan-cancer diagnosis (area under the receiver operating characteristic curve [AUROC] = 0.956), future cancer prediction up to 1 year before diagnosis (AUROC = 0.869), tumor stage inference (mean AUROC > 0.90), patient-specific treatment-response forecasting, and recurrence-free survival stratification across ten cancer types (all p < 0.01). Staging predictions were independently validated against postoperative pathological endpoints and shown to converge on core cancer genomic pathways. By translating routine clinical data into a dynamic view of cancer evolution, Oncoformer provides a framework for risk-informed cancer prediction and treatment stratification using routine clinical data.
PMID: 42508404
ISSN: 1097-4172
CID: 6070394

Prediction of maternal and infant outcomes from longitudinal electronic health records with a Mother-Child AI agent

Liu, Sian; Zheng, Wenxin; Kang, Jin; Xu, Tianyi; Chen, Siming; Li, Gen; Li, Junlong; Wong, Hang; Wang, Meihao; Bai, Xiaokai; Hu, Changxi; Tang, Cheng; Jin, Shengwei; Zou, Zixing; Chong, Ieng; Lu, Yuxing; Wong, Io Nam; Xu, Hui; Zhang, Charlotte L; Shi, Jingman; Feng, Erhu; Gu, Jinyu; Sun, Zhuo; Chen, Haibo; Yang, Li; Zhang, Yuan; Zhu, Xian; Huang, Huanhuan; Xu, Xiuyuan; Li, Xue; Zhenhui, Zhao; Qi, Hongbo; Lu, Xinyu; Cheng, Ngaman; Pan, Sicheng; Sun, Ning; Yin, Yun; Williams, Michelle; Oermann, Eric; Rasko, John E J; Li, Jin; Wang, Kai; Zhang, Kang; Wu, Hao; Xia, Yubin; Zeng, Fanxin; ,
Current predictive models for pregnancy and infant outcomes often focus on limited endpoints and rely on costly tests or imaging. Here we developed the Mother-Child AI Agent (MoChiAgent), an LLM-based clinical assistant that orchestrates multiple tools to integrate sequential electronic health record (EHR) data, including routine laboratory tests, for forecasting maternal and infant diseases. MoChiAgent's core predictive engine, MoChiFormer, was developed and internally evaluated using 4,401,599 longitudinal clinical visits and externally validated using independent maternal and infant cohorts consisting of 263,452 and 23,192 visits, respectively. MoChiFormer reconstructs missing laboratory values, reduces batch effects and learns EHR representations that support gestational, fetal and infant age estimation, health-trajectory modelling and stratification of current and future disease risk. Subsequently, a Knowledge Search Tool utilizes these forecasts to retrieve evidence-based intervention and treatment recommendations from curated medical literature and authoritative guidelines. For maternal health, MoChiFormer accurately identified key gestational conditions, achieving AUROCs of 0.89 for placental abruption, 0.89 for premature rupture of membranes, and 0.91 for preterm labour. Analysis of paired mother-infant data further revealed transgenerational risk associations, with infants born to mothers in specific clusters showing substantially elevated risks of neonatal jaundice (HR = 2.81, 95% CI 2.60-3.03) and haematological diseases (HR = 2.83, 95% CI 2.62-3.05). Integrating maternal gestational EHRs with infant records improved prediction of infant conditions, including chromosomal abnormalities and respiratory disorders. These findings suggest that MoChiAgent can provide clinically relevant, actionable decision-support information to enhance risk-stratified care for mothers and infants.
PMID: 42742183
ISSN: 1546-170x
CID: 6072883

Evaluating the Performance and Fragility of Large Language Models on the Self-Assessment for Neurological Surgeons

Vishwanath, Krithik; Alyakin, Anton; Ghosh, Mrigayu; Lee, Jin Vivian; Alber, Daniel Alexander; Sangwon, Karl L; Kondziolka, Douglas; Oermann, Eric Karl
BACKGROUND AND OBJECTIVES/OBJECTIVE:The Congress of Neurological Surgeons Self-Assessment for Neurological Surgeons questions are widely used by neurosurgical residents to prepare for written board examinations. Recently, these questions have also served as benchmarks for evaluating large language models' (LLMs) neurosurgical knowledge. LLMs show significant promise for transforming neurosurgical practice; however, they are susceptible to in-text distractions and confounding factors. Given the increasing use of generative artificial intelligence and ambient dictation technologies, clinical text is at a larger risk for the inclusion of extraneous details. The aim of this study was to assess the performance of state-of-the-art LLMs on neurosurgery board-like questions and to evaluate their robustness to the inclusion of distractor statements. METHODS:A comprehensive evaluation was conducted using 28 state-of-the-art LLMs. These models were tested on 2904 neurosurgery board examination questions derived from the Congress of Neurological Surgeons Self-Assessment for Neurological Surgeons. In addition, the study introduced a distraction framework to assess the fragility of these models. The framework incorporated simple, irrelevant distractor statements containing polysemous words with clinical meanings used in nonclinical contexts to determine the extent to which such distractions degrade model performance on standard medical benchmarks. RESULTS:Six of the 28 tested LLMs achieved board-passing outcomes, with the top-performing models scoring over 15.7% above the passing threshold. When exposed to distractions, accuracy across various model architectures was significantly reduced-by as much as 20.4%-with 1 model failing that had previously passed. Both general-purpose and medical open-source models experienced greater performance declines compared with proprietary variants when subjected to the added distractors. CONCLUSION/CONCLUSIONS:While current LLMs demonstrate an impressive ability to answer neurosurgery board-like examination questions, their performance is markedly vulnerable to extraneous, distracting information. These findings underscore the critical need for developing novel mitigation strategies aimed at bolstering LLM resilience against in-text distractions, particularly for safe and effective clinical deployment.
PMID: 41358748
ISSN: 1524-4040
CID: 5977102

Spinal meningiomas: histopathological grading using a benchmark radiomics model with notes on disease control

Palla, Adhith; Goff, Nicolas K; Perdikis, Blake; Khan, Hammad A; Belakhoua, Sarra; Grin, Eric A; Valliani, Aly; Patel, Roshni; Yang, Jonathan T; McFaline-Figueroa, J Ricardo; Lau, Darryl; Frempong-Boadu, Anthony; Oermann, Eric K; Laufer, Ilya
OBJECTIVE:Spinal meningiomas (SMs) are common primary spinal tumors for which surgery is considered the first-line treatment when safe and feasible. The ability to extrapolate the tumor grade from preoperative imaging may significantly inform early patient expectation-setting regarding recurrence. Building on radiomics studies in cranial meningiomas, the authors aimed to construct a benchmark radiomics model to preoperatively identify the histological grade of SMs. METHODS:Institutional surgical records from May 2012 to November 2025 were queried for pathology-confirmed meningiomas below the foramen magnum, with preoperative contrast-enhanced imaging available for segmentation. SMs were classified as low-grade (WHO grade 1) and high-grade (WHO grade 2 tumors and grade 1 tumors with atypia). Tumors were manually segmented, and features were extracted using the PyRadiomics software package. An ensemble model of k-nearest neighbors, random forest, and support vector machine classifiers was trained using nested cross-validation on a subset of 10 features to differentiate tumor grades. Clinical data for the cohort were also extracted, and disease control in an adjunctive clinical series was assessed. RESULTS:Seventy-four patients were included in radiomics analysis, with an area under the receiver operating characteristic curve of 0.879 and a mean F1 score of 0.748. The model's top 5 features were all texture features that differed significantly (p < 0.05) across low- and high-grade SMs. These included measures of tumor textural and contrast-enhancement heterogeneity, with overlap with features reported in radiomics models for histological grading of intracranial meningiomas. Fifty-five patients with a median radiographic follow-up of 22.2 (range 1.9-86.4) months remained for clinical analysis after exclusion of patients with less than 1 month of follow-up and syndromic meningiomas. Four recurrences occurred at a median of 20.8 (range 1.8-41.8) months. High-grade tumor pathology did not significantly impact progression-free survival (p = 0.682, log-rank test; Cox regression high vs low grade hazard ratio [HR] 0.62, 95% CI 0.06-6.11, p = 0.685). Subtotal resection was associated with poorer progression-free survival than gross-total resection (p = 0.004, log-rank test; Cox regression subtotal vs gross-total resection HR 10.62, 95% CI 1.46-77.05, p = 0.019). These findings remain contextualized within a relatively limited follow-up window and small recurrence event count, suggesting a need to characterize the interplay between tumor grade and extent of resection as drivers of local disease control in SMs. CONCLUSIONS:A preoperative radiomics model can stratify high-grade SMs using open-source tools applied to single-institution data.
PMID: 42679405
ISSN: 1092-0684
CID: 6071950

Longitudinal alignments and syntheses of multimodal clinical data for personalized medicine with the PULSE framework

Wu, Wei; Li, Gen; Wang, Kai; Xu, Hui; Xiao, Haodi; Hu, Changxi; Liu, Sian; Tang, Cheng; Liu, Fei; Zou, Zixing; Li, Bingzhou; Li, Jinghang; Zhang, Charlotte L; Wong, Hang; Chong, Ieng; Lu, Wenyang; Sun, Zhuo; Yin, Yun; Loupy, Alexandre; Oermann, Eric; Al Dajani, Saleem A; Zhu, Hao; Gootenberg, Jonathan; Abudayyeh, Omar O; Gladyshev, Vadim N; Rasko, John E J; Zhang, Kang; ,
Multimodal models capable of imputing diverse data used in single-cell biology studies provide potential foundational opportunities in clinical practice. However, patient data uniquely comprise longitudinal mosaic measurements that reflect underlying physiological dynamics and exhibit temporal covariation, demanding a specialized approach. Here we present Patient Unified Longitudinal Signal Engine (PULSE), a longitudinal self-supervised framework that explicitly encodes personalized past states (historical paired modalities) to reconstruct full profiles from subsequent unpaired measurements, thus enhancing current visit multimodal alignment and generation. Applied to the UK Biobank, PULSE accurately generates metabolomic profiles and proteomic profiles from sparse routine blood tests. Compared with the ground truth metabolomic data (251 biomarkers), PULSE-generated profiles outperformed all benchmark methods. Furthermore, the framework accommodates incorporation of retinal images, electronic health records and blood markers with disease prediction: models trained on the generated proteomic profiles achieved areas under the curve of 0.72-0.83 for six common diseases, comparable to that using ground-truth proteomic data. The PULSE framework demonstrates that cross-modal alignment captures the continuous spectrum of disease physiology and extracts robust features that transcend the limitations of traditional binary case-controls.
PMID: 42649447
ISSN: 2662-8457
CID: 6071810