Biomedical Informatics Research Research Articles

Large language models for biomedicine: foundations, opportunities, challenges, and best practices.

Generative large language models (LLMs) are a subset of transformers-based neural network architecture models. LLMs have successfully leveraged a combination of an increased number of parameters, improvements in computational efficiency, and large pre-training datasets to perform a wide spectrum of natural language processing (NLP) tasks. Using a few examples (few-shot) or no examples (zero-shot) for prompt-tuning has enabled LLMs to achieve state-of-the-art performance in a broad range of NLP applications. This article by the American Medical Informatics Association (AMIA) NLP Working Group characterizes the opportunities, challenges, and best practices for our community to leverage and advance the integration of LLMs in downstream NLP applications effectively. This can be accomplished through a variety of approaches, including augmented prompting, instruction prompt tuning, and reinforcement learning from human feedback (RLHF). Our focus is on making LLMs accessible to the broader biomedical informatics community, including clinicians and researchers who may be unfamiliar with NLP. Additionally, NLP practitioners may gain insight from the described best practices. We focus on 3 broad categories of NLP tasks, namely natural language understanding, natural language inferencing, and natural language generation. We review the emerging trends in prompt tuning, instruction fine-tuning, and evaluation metrics used for LLMs while drawing attention to several issues that impact biomedical NLP applications, including falsehoods in generated text (confabulation/hallucinations), toxicity, and dataset contamination leading to overfitting. We also review potential approaches to address some of these current challenges in LLMs, such as chain of thought prompting, and the phenomena of emergent capabilities observed in LLMs that can be leveraged to address complex NLP challenge in biomedical applications.

Inaugural Editorial: Editorial about the Founding of Medinformatics

The convergence of biology and artificial intelligence represents a paradigm shift in scientific exploration. The integration of AI into the study of omics—genomics, proteomics, transcriptomics, metabolomics, and beyond—has revolutionized our ability to decipher the complexities of biological systems. From predictive modeling of gene structures to the elucidation of protein interactions, from the analysis of transcriptomic profiles to the exploration of epigenetic landscapes, AI has become an indispensable tool in the biologist's toolkit. In this journal that focuses on the integration of information and life medicine, we embark on countless journeys of artificial intelligence applications in biomedical fields, focusing on the application of omics analysis in analyzing biological processes or disease occurrence and development. In addition, this journal also focuses on the forefront of interdisciplinary research, with a particular focus on the knowledge and methods of artificial intelligence, mathematics, physics, and other disciplines to solve biomedical problems, enhancing our rational understanding of life and disease. As we founded this magazine, we invited researchers, scholars, and practitioners from different backgrounds to publish cutting-edge biomedical informatics research or join our editorial board. Let us explore the frontiers of knowledge, push the boundaries of innovation, and together, forge a brighter future for biomedical science.

Deep learning in bioinformatics.

Deep learning is a powerful machine learning technique that can learn from large amounts of data using multiple layers of artificial neural networks. This paper reviews some applications of deep learning in bioinformatics, a field that deals with analyzing and interpreting biological data. We first introduce the basic concepts of deep learning and then survey the recent advances and challenges of applying deep learning to various bioinformatics problems, such as genome sequencing, gene expression analysis, protein structure prediction, drug discovery, and disease diagnosis. We also discuss future directions and opportunities for deep learning in bioinformatics. We aim to provide an overview of deep learning so that bioinformaticians applying deep learning models can consider all critical technical and ethical aspects. Thus, our target audience is biomedical informatics researchers who use deep learning models for inference. This review will inspire more bioinformatics researchers to adopt deep-learning methods for their research questions while considering fairness, potential biases, explainability, and accountability.

A compressed large language model embedding dataset of ICD 10 CM descriptions

This paper presents novel datasets providing numerical representations of ICD-10-CM codes by generating description embeddings using a large language model followed by a dimension reduction via autoencoder. The embeddings serve as informative input features for machine learning models by capturing relationships among categories and preserving inherent context information. The model generating the data was validated in two ways. First, the dimension reduction was validated using an autoencoder, and secondly, a supervised model was created to estimate the ICD-10-CM hierarchical categories. Results show that the dimension of the data can be reduced to as few as 10 dimensions while maintaining the ability to reproduce the original embeddings, with the fidelity decreasing as the reduced-dimension representation decreases. Multiple compression levels are provided, allowing users to choose as per their requirements, download and use without any other setup. The readily available datasets of ICD-10-CM codes are anticipated to be highly valuable for researchers in biomedical informatics, enabling more advanced analyses in the field. This approach has the potential to significantly improve the utility of ICD-10-CM codes in the biomedical domain.

An Ontology-Based Approach to Improving Medication Appropriateness in Older Patients: Algorithm Development and Validation Study.

Background: Inappropriate medication in older patients with multimorbidity results in a greater risk of adverse drug events. Clinical decision support systems (CDSSs) are intended to improve medication appropriateness. One approach to improving CDSSs is to use ontologies instead of relational databases. Previously, we developed OntoPharma-an ontology-based CDSS for reducing medication prescribing errors. Objective: The primary aim was to model a domain for improving medication appropriateness in older patients (chronic patient domain). The secondary aim was to implement the version of OntoPharma containing the chronic patient domain in a hospital setting. Methods: A 4-step process was proposed. The first step was defining the domain scope. The chronic patient domain focused on improving medication appropriateness in older patients. A group of experts selected the following three use cases: medication regimen complexity, anticholinergic and sedative drug burden, and the presence of triggers for identifying possible adverse events. The second step was domain model representation. The implementation was conducted by medical informatics specialists and clinical pharmacists using Protégé-OWL (Stanford Center for Biomedical Informatics Research). The third step was OntoPharma-driven alert module adaptation. We reused the existing framework based on SPARQL to query ontologies. The fourth step was implementing the version of OntoPharma containing the chronic patient domain in a hospital setting. Alerts generated from July to September 2022 were analyzed. Results: We proposed 6 new classes and 5 new properties, introducing the necessary changes in the ontologies previously created. An alert is shown if the Medication Regimen Complexity Index is ≥40, if the Drug Burden Index is ≥1, or if there is a trigger based on an abnormal laboratory value. A total of 364 alerts were generated for 107 patients; 154 (42.3%) alerts were accepted. Conclusions: We proposed an ontology-based approach to provide support for improving medication appropriateness in older patients with multimorbidity in a scalable, sustainable, and reusable way. The chronic patient domain was built based on our previous research, reusing the existing framework. OntoPharma has been implemented in clinical practice and generates alerts, considering the following use cases: medication regimen complexity, anticholinergic and sedative drug burden, and the presence of triggers for identifying possible adverse events.

Large Language Models and Their Implications on Medical Education.

1Fourth-year medical student, Stanford University School of Medicine, Stanford, California; email: [email protected]; ORCID: https://orcid.org/0000-0002-3422-0373. 2Adjunct professor, Center for Biomedical Informatics Research, Stanford University School of Medicine, Stanford, California. Acknowledgements: The authors wish to thank Charles Prober, MD (Stanford University School of Medicine) for his assistance in reviewing the manuscript. Funding/Support: None reported. Other disclosures: None reported. Ethical approval: Reported as not applicable. First published online.

The IHI Rochester Report 2022 on Healthcare Informatics Research: Resuming After the CoViD-19

In 2020, the CoViD-19 pandemic spread worldwide in an unexpected way and suddenly modified many life issues, including social habits, social relationships, teaching modalities, and more. Such changes were also observable in many different healthcare and medical contexts. Moreover, the CoViD-19 pandemic acted as a stress test for many research endeavors, and revealed some limitations, especially in contexts where research results had an immediate impact on the social and healthcare habits of millions of people. As a result, the research community is called to perform a deep analysis of the steps already taken, and to re-think steps for the near and far future to capitalize on the lessons learned due to the pandemic. In this direction, on June 09th–11th, 2022, a group of twelve healthcare informatics researchers met in Rochester, MN, USA. This meeting was initiated by the Institute for Healthcare Informatics—IHI, and hosted by the Mayo Clinic. The goal of the meeting was to discuss and propose a research agenda for biomedical and health informatics for the next decade, in light of the changes and the lessons learned from the CoViD-19 pandemic. This article reports the main topics discussed and the conclusions reached. The intended readers of this paper, besides the biomedical and health informatics research community, are all those stakeholders in academia, industry, and government, who could benefit from the new research findings in biomedical and health informatics research. Indeed, research directions and social and policy implications are the main focus of the research agenda we propose, according to three levels: the care of individuals, the healthcare system view, and the population view.

Guest Editorial Innovations in Wearable, Implantable, Mobile, & Remote Healthcare With IoT & Sensor Informatics and Patient Monitoring

The papers presented in this special issue focus on technological innovations in wearable, implantable, mobile, and remote healthcare that include Internet of Things (IoT), sensor informatics, and patient monitoring applications. These new technologies are used to track the key signs of people’s health to improve their lifestyle and health disorders. Innovations in IoT devices play a vital role in assisting patients in managing their health conditions. Thanks to the advent of modern communication technologies and Internet of Things (IoT) paradigms that have made the implementation of biomedical devices nearly universal. Now with the evolving industrial revolution, patients and healthcare providers are expecting something more. Practically speaking, healthcare wearables have experienced tremendous growth in the past few years, and it is expected to grow even more shortly, making it an ideal space for the biomedical informatics research community to solve complex healthcare problems and more informed decision making to improve human health.

Using Recurrent Neural Networks for Predicting Type-2 Diabetes from Genomic and Tabular Data.

The development of genomic technology for smart diagnosis and therapies for various diseases has lately been the most demanding area for computer-aided diagnostic and treatment research. Exponential breakthroughs in artificial intelligence and machine intelligence technologies could pave the way for identifying challenges afflicting the healthcare industry. Genomics is paving the way for predicting future illnesses, including cancer, Alzheimer's disease, and diabetes. Machine learning advancements have expedited the pace of biomedical informatics research and inspired new branches of computational biology. Furthermore, knowing gene relationships has resulted in developing more accurate models that can effectively detect patterns in vast volumes of data, making classification models important in various domains. Recurrent Neural Network models have a memory that allows them to quickly remember knowledge from previous cycles and process genetic data. The present work focuses on type 2 diabetes prediction using gene sequences derived from genomic DNA fragments through automated feature selection and feature extraction procedures for matching gene patterns with training data. The suggested model was tested using tabular data to predict type 2 diabetes based on several parameters. The performance of neural networks incorporating Recurrent Neural Network (RNN) components, Long Short-Term Memory (LSTM), and Gated Recurrent Units (GRU) was tested in this research. The model's efficiency is assessed using the evaluation metrics such as Sensitivity, Specificity, Accuracy, F1-Score, and Mathews Correlation Coefficient (MCC). The suggested technique predicted future illnesses with fair Accuracy. Furthermore, our research showed that the suggested model could be used in real-world scenarios and that input risk variables from an end-user Android application could be kept and evaluated on a secure remote server.

A Novel 3D Unsupervised Domain Adaptation Framework for Cross-Modality Medical Image Segmentation.

We consider the problem of volumetric (3D) unsupervised domain adaptation (UDA) in cross-modality medical image segmentation, aiming to perform segmentation on the unannotated target domain (e.g. MRI) with the help of labeled source domain (e.g. CT). Previous UDA methods in medical image analysis usually suffer from two challenges: 1) they focus on processing and analyzing data at 2D level only, thus missing semantic information from the depth level; 2) one-to-one mapping is adopted during the style-transfer process, leading to insufficient alignment in the target domain. Different from the existing methods, in our work, we conduct a first of its kind investigation on multi-style image translation for complete image alignment to alleviate the domain shift problem, and also introduce 3D segmentation in domain adaptation tasks to maintain semantic consistency at the depth level. In particular, we develop an unsupervised domain adaptation framework incorporating a novel quartet self-attention module to efficiently enhance relationships between widely separated features in spatial regions on a higher dimension, leading to a substantial improvement in segmentation accuracy in the unlabeled target domain. In two challenging cross-modality tasks, specifically brain structures and multi-organ abdominal segmentation, our model is shown to outperform current state-of-the-art methods by a significant margin, demonstrating its potential as a benchmark resource for the biomedical and health informatics research community.

About The Author

About The Author

Comprehension, utility, and preferences of prostate cancer survivors for visual timelines of patient-reported outcomes co-designed for limited graph literacy: meters and emojis over comics.

Visual timelines of patient-reported outcomes (PRO) can help prostate cancer survivors manage longitudinal data, compare with population averages, and consider future trajectories. PRO visualizations are most effective when designed with deliberate consideration of users. Yet, graph literacy is often overlooked as a design constraint, particularly when users with limited graph literacy are not engaged in their development. We conducted user testing to assess comprehension, utility, and preference of longitudinal PRO visualizations designed for prostate cancer survivors with limited literacy. Building upon our prior work co-designing longitudinal PRO visualizations with survivors, we engaged 18 prostate cancer survivors in a user study to assess 4 prototypes: Meter, Words, Comic, and Emoji. During remote sessions, we collected data on prototype comprehension (gist and verbatim), utility, and preference. Participants were aged 61-77 (M = 69), of whom half were African American. The majority of participants had less than a college degree (95%), had inadequate health literacy (78%), and low graph literacy (89%). Among the 4 prototypes, Meter had the best gist comprehension and was preferred. Emoji was also preferred, had the highest verbatim comprehension, and highest rated utility, including helpfulness, confidence, and satisfaction. Meter and Words both rated mid-range for utility, and Words scored lower than Emoji and Meter for comprehension. Comic had the poorest comprehension, lowest utility, and was least preferred. Findings identify design considerations for PRO visualizations, contributing to the knowledge base for visualization best practices. We describe our process to meaningfully engage patients from diverse and hard-to-reach groups for remote user testing, an important endeavor for health equity in biomedical informatics. Graph literacy is an important design consideration for PRO visualizations. Biomedical informatics researchers should be intentional in understanding user needs by involving diverse and representative individuals during development.

AI in NSCLC: PET-CT & histology model.

e21044 Background: Almost 70% of cases of lung cancer are diagnosed at advanced stage, albeit with sophisticated imaging modalities available. Of these, 35% are tiny nodules that are often missed at initial radiological screening, owing to limitations of resolution of human vision.Dramatic shift in therapeutic paradigm in the management of lung cancer in the last decade has ushered an era of stratified medicine. In India, one-third population inhabits rural lands;lack of awareness for molecular medicine poses challenges to treating oncologists. AI-based cancer detection and prediction of potential oncogenic driver at diagnosis, will help alleviate patient anxiety caused due to diagnostic delays, aiding in prompt initiation of optimal therapy.Here, a comprehensive deep-CNN based diagnosis of lung adenocarcinoma (LUAD) was implemented using both PET/CT and histopathology images of 100 patients. Methods: Data collection: 143 whole-slide images were downloaded from Dartmouth Lung Cancer Histology Dataset from Biomedical Informatics Research and Data Science website and PET/CT dataset-A Large-Scale CT and PET/CT Dataset for Lung Cancer Diagnosis (Lung-PET-CT-Dx) from The Cancer Imaging Archive. This comprises 251,135 images from 355 participants across 436 studies. Creation of Deep-CNN model: Residual Neural Network - 18 was used for histopathlogic classification followed by Detectron2’s region-based deep-CNN for detection of presence of tumours in lung PET/CT images. Properties such as size, morphology, attenuation, malignancy, and spiculation were also measured. The 18-layer deep network, ResNet-18, detected the histologic patterns using whole-slide histopathology images. Results: Results for 100 patients whose PET/CT images, histopathology images, and mutational status of three genes (EGFR, ALK, ROS) were readily available. The histologic patterns of tumour cells appear to be related to the mutational status of genes. The ResNet-18 CNN could predict various histologic patterns (Acinar, Lepidic, Papillary, Micropapillary, and Solid patterns). Conclusions: CNN models can help in early diagnosis and prompt treatment of lung cancer, if fairly accurate. Larger datasets are needed to improve the same. The final validation of the model is underway and will be reported in near future.[Table: see text]

NLM's sponsorship of research in biomedical informatics (1985-2016).

The U.S. National Library of Medicine’s (NLM) funding for biomedical informatics research in the 1980s and 1990s focused on clinical decision support systems, which were also the focus of research for Donald A.B. Lindberg M.D. prior to becoming NLM’s director. The portfolio of projects expanded over the years. At NLM, Dr. Lindberg supported various large infrastructure programs that enabled biomedical informatics research, as well as investigator-initiated research projects that increasingly included biotechnology/bioinformatics and health services research. The authors review NLM’s sponsorship of research during Dr. Lindberg’s tenure as its Director. NLM’s funding significantly increased in the 2000’s and beyond. Authors report an analysis of R01 topics from 1985–2016 using data from NIH RePORTER. Dr. Lindberg’s legacy for biomedical informatics research is reflected by the research NLM supported under his leadership. The number of R01s remained steady over the years, but the funds provided within awards increased over time. A significant amount of NLM funds listed in RePORTER went into various types of infrastructure projects that laid a solid foundation for biomedical informatics research over multiple decades.

Don Lindberg's influence on future generations: The U.S. National Library of Medicine's biomedical informatics research training programs.

Through his visionary leadership as Director of the U.S. National Library of Medicine (NLM), Donald A. B. Lindberg M.D. influenced future generations of informatics professionals and the field of biomedical informatics itself. This chapter describes Dr. Lindberg’s role in sponsoring and shaping the NLM’s Institutional T15 training programs.

NLM's Sponsorship of Research in Biomedical Informatics (1985-2016).

The U.S. National Library of Medicine's (NLM) funding for biomedical informatics research in the 1980s and 1990s focused on clinical decision support systems, which were also the focus of research for Donald A.B. Lindberg M.D. prior to becoming NLM's director. The portfolio of projects expanded over the years. At NLM, Dr. Lindberg supported various large infrastructure programs that enabled biomedical informatics research, as well as investigator-initiated research projects that increasingly included biotechnology/bioinformatics and health services research. The authors review NLM's sponsorship of research during Dr. Lindberg's tenure as its Director. NLM's funding significantly increased in the 2000's and beyond. Authors report an analysis of R01 topics from 1985-2016 using data from NIH RePORTER. Dr. Lindberg's legacy for biomedical informatics research is reflected by the research NLM supported under his leadership. The number of R01s remained steady over the years, but the funds provided within awards increased over time. A significant amount of NLM funds listed in RePORTER went into various types of infrastructure projects that laid a solid foundation for biomedical informatics research over multiple decades.

DNA Sequence Analysis for Brain Disorder Using Deep Learning and Secure Storage

Analysis of brain disorder in the neuroimaging of Magnetic Resonance Imaging (MRI), Positron Emission Tomography (PET), and Computed Tomography (CT) needs to understand the functionalities of the brain and it has been performed using traditional methods. Deep learning algorithms have also been applied in genomics data processing. The brain disorder diseases of Alzheimer, Schizophrenia, and Parkinson are analyzed in this work. The main issue in the traditional algorithm is the improper detection of disorders in the neuroimaging data. This paper presents a deep learning algorithm for the classification of brain disorder using Deep Belief Network (DBN) and securely storing the image using a Deoxyribonucleic Acid (DNA) Sequence-based Joint Photographic Experts Group (JPEG) Zig Zag Encryption Algorithm (DBNJZZ). In this work, DBNJZZ implements an efficient and effective prediction model for disorders using the open-access datasets of Alzheimer's Disease Neuroimaging Initiative (Adni), the Center for Biomedical Research Excellence (Cobre), the Open Access Series of Imaging Studies (Oasis), the Function Biomedical Informatics Research Network (Fbirn), a Parkinson's dataset of 55 patients and 23 subjects with Parkinson's syndromes (Ntua), and the Parkinson's Progression Markers Initiative (Ppmi). This algorithm is implemented and tested using performance metric measures of accuracy, Root Mean Square Error (RMSE), Mean Absolute Error (MAE), and Mean Absolute Percentage Error (MAPE). DBNJZZ gives better performance with an accuracy of 99.21% and also surpasses previous methods on other measures.

Translating Data Analytics Into Improved Spine Surgery Outcomes: A Roadmap for Biomedical Informatics Research in 2021.

Study Design:Narrative review.Objectives:There is growing interest in the use of biomedical informatics and data analytics tools in spine surgery. Yet despite the rapid growth in research on these topics, few analytic tools have been implemented in routine spine practice. The purpose of this review is to provide a health information technology (HIT) roadmap to help translate data assets and analytics tools into measurable advances in spine surgical care.Methods:We conducted a narrative review of PubMed and Google Scholar to identify publications discussing data assets, analytical approaches, and implementation strategies relevant to spine surgery practice.Results:A variety of data assets are available for spine research, ranging from commonly used datasets, such as administrative billing data, to emerging resources, such as mobile health and biobanks. Both regression and machine learning techniques are valuable for analyzing these assets, and researchers should recognize the particular strengths and weaknesses of each approach. Few studies have focused on the implementation of HIT, and a variety of methods exist to help translate analytic tools into clinically useful interventions. Finally, a number of HIT-related challenges must be recognized and addressed, including stakeholder acceptance, regulatory oversight, and ethical considerations.Conclusions:Biomedical informatics has the potential to support the development of new HIT that can improve spine surgery quality and outcomes. By understanding the development life-cycle that includes identifying an appropriate data asset, selecting an analytic approach, and leveraging an effective implementation strategy, spine researchers can translate this potential into measurable advances in patient care.

Why Is the Electronic Health Record So Challenging for Research and Clinical Care?

The electronic health record (EHR) has become increasingly ubiquitous. At the same time, health professionals have been turning to this resource for access to data that is needed for the delivery of health care and for clinical research. There is little doubt that the EHR has made both of these functions easier than earlier days when we relied on paper-based clinical records. Coupled with modern database and data warehouse systems, high-speed networks, and the ability to share clinical data with others are large number of challenges that arguably limit the optimal use of the EHR OBJECTIVES: Our goal was to provide an exhaustive reference for those who use the EHR in clinical and research contexts, but also for health information systems professionals as they design, implement, and maintain EHR systems. This study includes a panel of 24 biomedical informatics researchers, information technology professionals, and clinicians, all of whom have extensive experience in design, implementation, and maintenance of EHR systems, or in using the EHR as clinicians or researchers. All members of the panel are affiliated with Penn Medicine at the University of Pennsylvania and have experience with a variety of different EHR platforms and systems and how they have evolved over time. Each of the authors has shared their knowledge and experience in using the EHR in a suite of 20 short essays, each representing a specific challenge and classified according to a functional hierarchy of interlocking facets such as usability and usefulness, data quality, standards, governance, data integration, clinical care, and clinical research. We provide here a set of perspectives on the challenges posed by the EHR to clinical and research users.

Optimal Fuzzy Cluster Partitioning by Crow Search Meta-Heuristic for Biomedical Data Analysis

Biomedical data is often more unstructured in nature, and biomedical data processing task is becoming more complex day by day. Thus, biomedical informatics requires competent data analysis and data mining techniques for designing decision support system's framework to solve clinical and heathcare-related issues. Due to increasingly large and complex data sets and demand of biomedical informatics research, researchers are attracted towards automated machine learning models. This paper is proposed to design an efficient machine learning model based on fuzzy c-means with meta-heuristic optimizations for biomedical data analysis and clustering. The main contributions of this paper are 1) projecting an efficient machine learning model based on fuzzy c-means and meta-heuristic optimization for biomedical data classification, 2) employing benchmark validation techniques and critical hypothesises testing, and 3) providing a background for biomedical data processing with a view of data processing and mining.