Anyi Li
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Anyi Li has a total of 137 co-authors. Here are the top co-authors by number of shared publications:
- Hui Lin
Department of Radiation Oncology, University of California San Francisco, San Francisco, USA
2 shared publications Karl Pichotta
Memorial Sloan Kettering Cancer Center, New York, USA
2 shared publications- Lior Z. Braunstein
Department of Radiation Oncology, Memorial Sloan Kettering Cancer Center, New York, NY
2 shared publications Menglei Zhu
Department of Pathology and Laboratory Medicine Memorial Sloan Kettering Cancer Center
2 shared publications- Michele Waters
Computational Oncology Service, Department of Epidemiology and Biostatistics, Memorial Sloan Kettering Cancer Center, New York, USA
2 shared publications Sulov Chalise
Department of Pathology and Laboratory Medicine Memorial Sloan Kettering Cancer Center
2 shared publications
Latest Publications
- Open Access
- 22 September 2026
- Published by Springer Nature in npj Digital Medicine
Leveraging dynamic prompting for outcome prediction of cancer patients using large language models and electronic health record notes
Outcome prediction from unstructured EHR notes remains challenging, especially for rare cancers with limited representation in large language model (LLM) pre-training data. We present a dynamic prompting framework that retrieves semantically similar patient examples, constructs tailored few-shot prompts and integrates note summaries to enhance outcome predictions. We evaluated this approach on a retrospective cohort of 503 breast cancer and 475 glioma patients using EHR notes from the first 180 days post-diagnosis. Dynamic prompting substantially improved glioma prediction while producing modest improvements for breast cancer, consistent with our hypothesis that dynamic prompting benefits rare cancers more. The combined summarization-driven dynamic-prompting approach achieved the highest performance while maintaining stability. Embedding visualizations suggested alignment with established prognostic markers in both cohorts. In this proof-of-concept, dynamic prompting improved outcome prediction for a lower-incidence cancer (glioma) in a single cohort while maintaining performance for a common cancer (breast), establishing it as a promising approach for LLM-based outcome prediction from unstructured EHR notes. Validation across additional low-incidence tumor types and institutions is required before these findings can be generalized.
- 22 September 2026
- Published by Springer Nature in npj Digital Medicine
Open AccessLeveraging dynamic prompting for outcome prediction of cancer patients using large language models and electronic health record notes
Outcome prediction from unstructured EHR notes remains challenging, especially for rare cancers with limited representation in large language model (LLM) pre-training data. We present a dynamic prompting framework that retrieves semantically similar patient examples, constructs tailored few-shot prompts and integrates note summaries to enhance outcome predictions. We evaluated this approach on a retrospective cohort of 503 breast cancer and 475 glioma patients using EHR notes from the first 180 days post-diagnosis. Dynamic prompting substantially improved glioma prediction while producing modest improvements for breast cancer, consistent with our hypothesis that dynamic prompting benefits rare cancers more. The combined summarization-driven dynamic-prompting approach achieved the highest performance while maintaining stability. Embedding visualizations suggested alignment with established prognostic markers in both cohorts. In this proof-of-concept, dynamic prompting improved outcome prediction for a lower-incidence cancer (glioma) in a single cohort while maintaining performance for a common cancer (breast), establishing it as a promising approach for LLM-based outcome prediction from unstructured EHR notes. Validation across additional low-incidence tumor types and institutions is required before these findings can be generalized.
- Open Access
- 16 February 2026
- Published by Springer Nature in npj Digital Medicine
- Vol. 9 (1)
Artificial intelligence-based incident analysis and learning system to enhance patient safety and improve treatment quality
Patient safety and high treatment quality are essential in modern healthcare, but analyzing safety incidents for insights is labor-intensive and inconsistent process. To address this, we developed the Artificial Intelligence-based Incident Analysis and Learning System (AI-ILS), trained on 1548 expertly curated incidents categorized by the Human Factors Analysis and Classification System (HFACS). AI-ILS identifies latent safety threats and classifies incident causes with high accuracy, achieving an average AUROC of 0.92, MCC of 0.72, and overall accuracy of 79%. In testing on 350 real-world clinical incidents, AI-ILS showed 88% concordance with expert reviewers and operated 29 times faster than manual analysis. We deployed and validated AI-ILS using real-world radiation oncology data, where it improved retrospective incident analysis at our institution by generating aggregated HFACS-based results and addressing challenges related to inconsistent review processes and lack of standardized taxonomies.
- 16 February 2026
- Published by Springer Nature in npj Digital Medicine
- Vol. 9 (1)
Open AccessArtificial intelligence-based incident analysis and learning system to enhance patient safety and improve treatment quality
Patient safety and high treatment quality are essential in modern healthcare, but analyzing safety incidents for insights is labor-intensive and inconsistent process. To address this, we developed the Artificial Intelligence-based Incident Analysis and Learning System (AI-ILS), trained on 1548 expertly curated incidents categorized by the Human Factors Analysis and Classification System (HFACS). AI-ILS identifies latent safety threats and classifies incident causes with high accuracy, achieving an average AUROC of 0.92, MCC of 0.72, and overall accuracy of 79%. In testing on 350 real-world clinical incidents, AI-ILS showed 88% concordance with expert reviewers and operated 29 times faster than manual analysis. We deployed and validated AI-ILS using real-world radiation oncology data, where it improved retrospective incident analysis at our institution by generating aggregated HFACS-based results and addressing challenges related to inconsistent review processes and lack of standardized taxonomies.
- Open Access
AI-assisted workflow enables rapid, high-fidelity breast cancer clinical trial eligibility prescreening
Clinical trials play an important role in cancer care and research, yet participation rates remain low. We developed MSK-MATCH (Memorial Sloan Kettering Multi-Agent Trial Coordination Hub), an AI system for automated eligibility screening from clinical text. MSK-MATCH integrates a large language model with a curated oncology trial knowledge base and retrieval-augmented architecture providing explanations for all AI predictions grounded in source text. In a retrospective dataset of 88,518 clinical documents from 731 patients across six breast cancer trials, MSK-MATCH automatically resolved 61.9% of cases and triaged 38.1% for human review. This AI-assisted workflow achieved 98.6% accuracy, 98.4% sensitivity, and 98.7% specificity for patient-level eligibility classification, matching or exceeding performance of the human-only and AI-only comparisons. For the triaged cases requiring manual review, prepopulating eligibility screens with AI-generated explanations reduced screening time from 20 minutes to 43 seconds at an average cost of $0.96 per patient-trial pair.
Open AccessAI-assisted workflow enables rapid, high-fidelity breast cancer clinical trial eligibility prescreening
Clinical trials play an important role in cancer care and research, yet participation rates remain low. We developed MSK-MATCH (Memorial Sloan Kettering Multi-Agent Trial Coordination Hub), an AI system for automated eligibility screening from clinical text. MSK-MATCH integrates a large language model with a curated oncology trial knowledge base and retrieval-augmented architecture providing explanations for all AI predictions grounded in source text. In a retrospective dataset of 88,518 clinical documents from 731 patients across six breast cancer trials, MSK-MATCH automatically resolved 61.9% of cases and triaged 38.1% for human review. This AI-assisted workflow achieved 98.6% accuracy, 98.4% sensitivity, and 98.7% specificity for patient-level eligibility classification, matching or exceeding performance of the human-only and AI-only comparisons. For the triaged cases requiring manual review, prepopulating eligibility screens with AI-generated explanations reduced screening time from 20 minutes to 43 seconds at an average cost of $0.96 per patient-trial pair.
- Top Cited PapersOpen Access
- 2 July 2025
- Published by Springer Nature in npj Digital Medicine
- Vol. 8 (1) , 1-15
Large language model trained on clinical oncology data predicts cancer progression
Subspecialty knowledge barriers have limited the adoption of large language models (LLMs) in oncology. We introduce Woollie, an open-source, oncology-specific LLM trained on real-world data from Memorial Sloan Kettering Cancer Center (MSK) across lung, breast, prostate, pancreatic, and colorectal cancers, with external validation using University of California, San Francisco (UCSF) data. Woollie surpasses ChatGPT in medical benchmarks and excels in eight non-medical benchmarks. Analyzing 39,319 radiology impression notes from 4002 patients, it achieved an overall area under the receiver operating characteristic curve (AUROC) of 0.97 for cancer progression prediction on MSK data, including a notable 0.98 AUROC for pancreatic cancer. On UCSF data, it achieved an overall AUROC of 0.88, excelling in lung cancer detection with an AUROC of 0.95. As the first oncology specific LLM validated across institutions, Woollie demonstrates high accuracy and consistency across cancer types, underscoring its potential to enhance cancer progression analysis.
- 2 July 2025
- Published by Springer Nature in npj Digital Medicine
- Vol. 8 (1) , 1-15
Top Cited PapersOpen AccessLarge language model trained on clinical oncology data predicts cancer progression
Subspecialty knowledge barriers have limited the adoption of large language models (LLMs) in oncology. We introduce Woollie, an open-source, oncology-specific LLM trained on real-world data from Memorial Sloan Kettering Cancer Center (MSK) across lung, breast, prostate, pancreatic, and colorectal cancers, with external validation using University of California, San Francisco (UCSF) data. Woollie surpasses ChatGPT in medical benchmarks and excels in eight non-medical benchmarks. Analyzing 39,319 radiology impression notes from 4002 patients, it achieved an overall area under the receiver operating characteristic curve (AUROC) of 0.97 for cancer progression prediction on MSK data, including a notable 0.98 AUROC for pancreatic cancer. On UCSF data, it achieved an overall AUROC of 0.88, excelling in lung cancer detection with an AUROC of 0.95. As the first oncology specific LLM validated across institutions, Woollie demonstrates high accuracy and consistency across cancer types, underscoring its potential to enhance cancer progression analysis.
- Top Cited PapersOpen Access
- 6 November 2024
- Published by Springer Nature in Nature
- Vol. 636 (8043) , 728-736
Automated real-world data integration improves cancer outcome prediction
The digitization of health records and growing availability of tumour DNA sequencing provide an opportunity to study the determinants of cancer outcomes with unprecedented richness. Patient data are often stored in unstructured text and siloed datasets. Here we combine natural language processing annotations1,2 with structured medication, patient-reported demographic, tumour registry and tumour genomic data from 24,950 patients at Memorial Sloan Kettering Cancer Center to generate a clinicogenomic, harmonized oncologic real-world dataset (MSK-CHORD). MSK-CHORD includes data for non-small-cell lung (n = 7,809), breast (n = 5,368), colorectal (n = 5,543), prostate (n = 3,211) and pancreatic (n = 3,109) cancers and enables discovery of clinicogenomic relationships not apparent in smaller datasets. Leveraging MSK-CHORD to train machine learning models to predict overall survival, we find that models including features derived from natural language processing, such as sites of disease, outperform those based on genomic data or stage alone as tested by cross-validation and an external, multi-institution dataset. By annotating 705,241 radiology reports, MSK-CHORD also uncovers predictors of metastasis to specific organ sites, including a relationship between SETD2 mutation and lower metastatic potential in immunotherapy-treated lung adenocarcinoma corroborated in independent datasets. We demonstrate the feasibility of automated annotation from unstructured notes and its utility in predicting patient outcomes. The resulting data are provided as a public resource for real-world oncologic research.
- 6 November 2024
- Published by Springer Nature in Nature
- Vol. 636 (8043) , 728-736
Top Cited PapersOpen AccessAutomated real-world data integration improves cancer outcome prediction
The digitization of health records and growing availability of tumour DNA sequencing provide an opportunity to study the determinants of cancer outcomes with unprecedented richness. Patient data are often stored in unstructured text and siloed datasets. Here we combine natural language processing annotations1,2 with structured medication, patient-reported demographic, tumour registry and tumour genomic data from 24,950 patients at Memorial Sloan Kettering Cancer Center to generate a clinicogenomic, harmonized oncologic real-world dataset (MSK-CHORD). MSK-CHORD includes data for non-small-cell lung (n = 7,809), breast (n = 5,368), colorectal (n = 5,543), prostate (n = 3,211) and pancreatic (n = 3,109) cancers and enables discovery of clinicogenomic relationships not apparent in smaller datasets. Leveraging MSK-CHORD to train machine learning models to predict overall survival, we find that models including features derived from natural language processing, such as sites of disease, outperform those based on genomic data or stage alone as tested by cross-validation and an external, multi-institution dataset. By annotating 705,241 radiology reports, MSK-CHORD also uncovers predictors of metastasis to specific organ sites, including a relationship between SETD2 mutation and lower metastatic potential in immunotherapy-treated lung adenocarcinoma corroborated in independent datasets. We demonstrate the feasibility of automated annotation from unstructured notes and its utility in predicting patient outcomes. The resulting data are provided as a public resource for real-world oncologic research.