Hanyin Wang

Yikuan Li, Ramsey M. Wehbe, Faraz S. Ahmad et al.

Objective: Clinical knowledge enriched transformer models (e.g., ClinicalBERT) have state-of-the-art results on clinical NLP (natural language processing) tasks. One of the core limitations of these transformer models is the substantial memory consumption due to their full self-attention mechanism, which leads to the performance degradation in long clinical texts. To overcome this, we propose to leverage long-sequence transformer models (e.g., Longformer and BigBird), which extend the maximum input sequence length from 512 to 4096, to enhance the ability to model long-term dependencies in long clinical texts. Materials and Methods: Inspired by the success of long sequence transformer models and the fact that clinical notes are mostly long, we introduce two domain enriched language models, Clinical-Longformer and Clinical-BigBird, which are pre-trained on a large-scale clinical corpus. We evaluate both language models using 10 baseline tasks including named entity recognition, question answering, natural language inference, and document classification tasks. Results: The results demonstrate that Clinical-Longformer and Clinical-BigBird consistently and significantly outperform ClinicalBERT and other short-sequence transformers in all 10 downstream tasks and achieve new state-of-the-art results. Discussion: Our pre-trained language models provide the bedrock for clinical NLP using long texts. We have made our source code available at https://github.com/luoyuanlab/Clinical-Longformer, and the pre-trained models available for public download at: https://huggingface.co/yikuan8/Clinical-Longformer. Conclusion: This study demonstrates that clinical knowledge enriched long-sequence transformers are able to learn long-term dependencies in long clinical text. Our methods can also inspire the development of other domain-enriched long-sequence transformers.

Adrienne Kline, Hanyin Wang, Yikuan Li et al.

As machine learning and artificial intelligence are more frequently being leveraged to tackle problems in the health sector, there has been increased interest in utilizing them in clinical decision-support. This has historically been the case in single modal data such as electronic health record data. Attempts to improve prediction and resemble the multimodal nature of clinical expert decision-making this has been met in the computational field of machine learning by a fusion of disparate data. This review was conducted to summarize this field and identify topics ripe for future research. We conducted this review in accordance with the PRISMA (Preferred Reporting Items for Systematic reviews and Meta-Analyses) extension for Scoping Reviews to characterize multi-modal data fusion in health. We used a combination of content analysis and literature searches to establish search strings and databases of PubMed, Google Scholar, and IEEEXplore from 2011 to 2021. A final set of 125 articles were included in the analysis. The most common health areas utilizing multi-modal methods were neurology and oncology. However, there exist a wide breadth of current applications. The most common form of information fusion was early fusion. Notably, there was an improvement in predictive performance performing heterogeneous data fusion. Lacking from the papers were clear clinical deployment strategies and pursuit of FDA-approved tools. These findings provide a map of the current literature on multimodal data fusion as applied to health diagnosis/prognosis problems. Multi-modal machine learning, while more robust in its estimations over unimodal methods, has drawbacks in its scalability and the time-consuming nature of information concatenation.

Yikuan Li, Chengsheng Mao, Kaixuan Huang et al.

Scarcity of health care resources could result in the unavoidable consequence of rationing. For example, ventilators are often limited in supply, especially during public health emergencies or in resource-constrained health care settings, such as amid the pandemic of COVID-19. Currently, there is no universally accepted standard for health care resource allocation protocols, resulting in different governments prioritizing patients based on various criteria and heuristic-based protocols. In this study, we investigate the use of reinforcement learning for critical care resource allocation policy optimization to fairly and effectively ration resources. We propose a transformer-based deep Q-network to integrate the disease progression of individual patients and the interaction effects among patients during the critical care resource allocation. We aim to improve both fairness of allocation and overall patient outcomes. Our experiments demonstrate that our method significantly reduces excess deaths and achieves a more equitable distribution under different levels of ventilator shortage, when compared to existing severity-based and comorbidity-based methods in use by different governments. Our source code is included in the supplement and will be released on Github upon publication.

Hanyin Wang, Meghan R. Hutch, Yikuan Li et al.

Over 12 billion doses of COVID-19 vaccines have been administered at the time of writing. However, public perceptions of vaccines have been complex. We analyzed COVID-19 vaccine-related tweets to understand the evolving perceptions of COVID-19 vaccines. We finetuned a deep learning classifier using a state-of-the-art model, XLNet, to detect each tweet's sentiment automatically. We employed validated methods to extract the users' race or ethnicity, gender, age, and geographical locations from user profiles. Incorporating multiple data sources, we assessed the sentiment patterns among subpopulations and juxtaposed them against vaccine uptake data to unravel their interactive patterns. 11,211,672 COVID-19 vaccine-related tweets corresponding to 2,203,681 users over two years were analyzed. The finetuned model for sentiment classification yielded an accuracy of 0.92 on testing set. Users from various demographic groups demonstrated distinct patterns in sentiments towards COVID-19 vaccines. User sentiments became more positive over time, upon which we observed subsequent upswing in the population-level vaccine uptake. Surrounding dates where positive sentiments crest, we detected encouraging news or events regarding vaccine development and distribution. Positive sentiments in pregnancy-related tweets demonstrated a delayed pattern compared with trends in general population, with postponed vaccine uptake trends. Distinctive patterns across subpopulations suggest the need of tailored strategies. Global news and events profoundly involved in shaping users' thoughts on social media. Populations with additional concerns, such as pregnancy, demonstrated more substantial hesitancy since lack of timely recommendations. Feature analysis revealed hesitancies of various subpopulations stemmed from clinical trial logics, risks and complications, and urgency of scientific evidence.

Hanyin Wang, Chufan Gao, Christopher Dantona et al.

In the U.S. inpatient payment system, the Diagnosis-Related Group (DRG) is pivotal, but its assignment process is inefficient. The study introduces DRG-LLaMA, an advanced large language model (LLM) fine-tuned on clinical notes to enhance DRGs assignment. Utilizing LLaMA as the foundational model and optimizing it through Low-Rank Adaptation (LoRA) on 236,192 MIMIC-IV discharge summaries, our DRG-LLaMA-7B model exhibited a noteworthy macro-averaged F1 score of 0.327, a top-1 prediction accuracy of 52.0%, and a macro-averaged Area Under the Curve (AUC) of 0.986, with a maximum input token length of 512. This model surpassed the performance of prior leading models in DRG prediction, showing a relative improvement of 40.3% and 35.7% in macro-averaged F1 score compared to ClinicalBERT and CAML, respectively. Applied to base DRG and complication or comorbidity (CC)/major complication or comorbidity (MCC) prediction, DRG-LLaMA achieved a top-1 prediction accuracy of 67.8% and 67.5%, respectively. Additionally, our findings indicate that DRG-LLaMA's performance correlates with increased model parameters and input context lengths.

Yikuan Li, Hanyin Wang, Halid Yerebakan et al.

In this study, we investigated the ability of the large language model (LLM) to enhance healthcare data interoperability. We leveraged the LLM to convert clinical texts into their corresponding FHIR resources. Our experiments, conducted on 3,671 snippets of clinical text, demonstrated that the LLM not only streamlines the multi-step natural language processing and human calibration processes but also achieves an exceptional accuracy rate of over 90% in exact matches when compared to human annotations.

Hanyin Wang, Chufan Gao, Bolun Liu et al.

Proprietary Large Language Models (LLMs) such as GPT-4 and Gemini have demonstrated promising capabilities in clinical text summarization tasks. However, due to patient data privacy concerns and computational costs, many healthcare providers prefer using small, locally-hosted models over external generic LLMs. This study presents a comprehensive domain- and task-specific adaptation process for the open-source LLaMA-2 13 billion parameter model, enabling it to generate high-quality clinical notes from outpatient patient-doctor dialogues. Our process incorporates continued pretraining, supervised fine-tuning, and reinforcement learning from both AI and human feedback. We introduced a new approach, DistillDirect, for performing on-policy reinforcement learning with Gemini 1.0 Pro as the teacher model. Our resulting model, LLaMA-Clinic, can generate clinical notes comparable in quality to those authored by physicians. In a blinded physician reader study, the majority (92.8%) of individual evaluations rated the notes generated by LLaMA-Clinic as "acceptable" or higher across three criteria: real-world readiness, completeness, and accuracy. In the more challenging "Assessment and Plan" section, LLaMA-Clinic matched physician-authored notes in real-world readiness score. We highlight key considerations for future clinical note-generation tasks, emphasizing the importance of pre-defining a "best practice" note format, rather than relying on LLMs to determine this for clinical practice.

Yikuan Li, Ramsey M. Wehbe, Faraz S. Ahmad et al.

Transformers-based models, such as BERT, have dramatically improved the performance for various natural language processing tasks. The clinical knowledge enriched model, namely ClinicalBERT, also achieved state-of-the-art results when performed on clinical named entity recognition and natural language inference tasks. One of the core limitations of these transformers is the substantial memory consumption due to their full self-attention mechanism. To overcome this, long sequence transformer models, e.g. Longformer and BigBird, were proposed with the idea of sparse attention mechanism to reduce the memory usage from quadratic to the sequence length to a linear scale. These models extended the maximum input sequence length from 512 to 4096, which enhanced the ability of modeling long-term dependency and consequently achieved optimal results in a variety of tasks. Inspired by the success of these long sequence transformer models, we introduce two domain enriched language models, namely Clinical-Longformer and Clinical-BigBird, which are pre-trained from large-scale clinical corpora. We evaluate both pre-trained models using 10 baseline tasks including named entity recognition, question answering, and document classification tasks. The results demonstrate that Clinical-Longformer and Clinical-BigBird consistently and significantly outperform ClinicalBERT as well as other short-sequence transformers in all downstream tasks. We have made our source code available at [https://github.com/luoyuanlab/Clinical-Longformer] the pre-trained models available for public download at: [https://huggingface.co/yikuan8/Clinical-Longformer].

Zifeng Wang, Hanyin Wang, Benjamin Danek et al.

The integration of Large Language Models (LLMs) into medical applications has sparked widespread interest across the healthcare industry, from drug discovery and development to clinical decision support, assisting telemedicine, medical devices, and healthcare insurance applications. This perspective paper aims to discuss the inner workings of building LLM-powered medical AI applications and introduces a comprehensive framework for their development. We review existing literature and outline the unique challenges of applying LLMs in specialized medical contexts. Additionally, we introduce a three-step framework to organize medical LLM research activities: 1) Modeling: breaking down complex medical workflows into manageable steps for developing medical-specific models; 2) Optimization: optimizing the model performance with crafted prompts and integrating external knowledge and tools, and 3) System engineering: decomposing complex tasks into subtasks and leveraging human expertise for building medical AI applications. Furthermore, we offer a detailed use case playbook that describes various LLM-powered medical AI applications, such as optimizing clinical trial design, enhancing clinical decision support, and advancing medical imaging analysis. Finally, we discuss various challenges and considerations for building medical AI applications with LLMs, such as handling hallucination issues, data ownership and compliance, privacy, intellectual property considerations, compute cost, sustainability issues, and responsible AI requirements.

Hanyin Wang, Zhenbang Wu, Gururaj Kolar et al.

Diagnosis-Related Group (DRG) codes are essential for hospital reimbursement and operations but require labor-intensive assignment. Large Language Models (LLMs) struggle with DRG coding due to the out-of-distribution (OOD) nature of the task: pretraining corpora rarely contain private clinical or billing data. We introduce DRG-Sapphire, which uses large-scale reinforcement learning (RL) for automated DRG coding from clinical notes. Built on Qwen2.5-7B and trained with Group Relative Policy Optimization (GRPO) using rule-based rewards, DRG-Sapphire introduces a series of RL enhancements to address domain-specific challenges not seen in previous mathematical tasks. Our model achieves state-of-the-art accuracy on the MIMIC-IV benchmark and generates physician-validated reasoning for DRG assignments, significantly enhancing explainability. Our study further sheds light on broader challenges of applying RL to knowledge-intensive, OOD tasks. We observe that RL performance scales approximately linearly with the logarithm of the number of supervised fine-tuning (SFT) examples, suggesting that RL effectiveness is fundamentally constrained by the domain knowledge encoded in the base model. For OOD tasks like DRG coding, strong RL performance requires sufficient knowledge infusion prior to RL. Consequently, scaling SFT may be more effective and computationally efficient than scaling RL alone for such tasks.

Hanyin Wang, Chufan Gao, Qiping Xu et al.

Process-supervised reward models (PRMs) excel at providing step-by-step verification for large language model (LLM) outputs in domains like mathematics and coding. However, their application to fields lacking ground-truth answers, such as clinical note generation, poses significant challenges. We introduce a novel framework for training PRMs to deliver step-level reward signals for LLM-generated clinical notes. By precisely defining meaningful "steps," injecting realistic "errors" informed by domain expertise, and leveraging LLMs to generate process supervision data at scale, we overcome previous limitations. Our PRM, built on LLaMA-3.1 8B, consistently outperforms proprietary reasoning and non-reasoning models, achieving state-of-the-art performance on two key evaluations: (1) distinguishing gold-standard from error-containing samples with 98.8% accuracy, and (2) selecting physician-preferred clinical notes with 56.2% accuracy. We investigate critical components for effective PRM training, including optimal loss functions and data selection strategies, and present a comprehensive physician reader study identifying predictors of downstream Best-of-N performance. Our study sheds light on unlocking the potential of PRMs for diverse generative tasks across domains.

Hanyin Wang, Yikuan Li, Andrew Naidech et al.

Background Sepsis is one of the most life-threatening circumstances for critically ill patients in the US, while a standardized criteria for sepsis identification is still under development. Disparities in social determinants of sepsis patients can interfere with the risk prediction performances using machine learning. Methods Disparities in social determinants, including race, gender, marital status, insurance types and languages, among patients identified by six available sepsis criteria were revealed by forest plots. Sixteen machine learning classifiers were trained to predict in-hospital mortality for sepsis patients. The performance of the trained model was tested on the entire randomly conducted test set and each sub-population built based on each of the following social determinants: race, gender, marital status, insurance type, and language. Results We analyzed a total of 11,791 critical care patients from the MIMIC-III database. Within the population identified by each sepsis identification method, significant differences were observed among sub-populations regarding race, marital status, insurance type, and language. On the 5,783 sepsis patients identified by the Sepsis-3 criteria statistically significant performance decreases for mortality prediction were observed when applying the trained machine learning model on Asian and Hispanic patients. With pairwise comparison, we detected performance discrepancies in mortality prediction between Asian and White patients, Asians and patients of other races, as well as English-speaking and Spanish-speaking patients. Conclusions Disparities in proportions of patients identified by various sepsis criteria were detected among the different social determinant groups. To achieve accurate diagnosis, a versatile diagnostic system for sepsis is needed to overcome the social determinant disparities of patients.

Yikuan Li, Hanyin Wang, Yuan Luo

Joint image-text embedding extracted from medical images and associated contextual reports is the bedrock for most biomedical vision-and-language (V+L) tasks, including medical visual question answering, clinical image-text retrieval, clinical report auto-generation. In this study, we adopt four pre-trained V+L models: LXMERT, VisualBERT, UNIER and PixelBERT to learn multimodal representation from MIMIC-CXR radiographs and associated reports. The extrinsic evaluation on OpenI dataset shows that in comparison to the pioneering CNN-RNN model, the joint embedding learned by pre-trained V+L models demonstrate performance improvement in the thoracic findings classification task. We conduct an ablation study to analyze the contribution of certain model components and validate the advantage of joint embedding over text-only embedding. We also visualize attention maps to illustrate the attention mechanism of V+L models.