Large Language Models in Medicine: The Potentials and Pitfalls : A Narrative Review.

Large language models (LLMs) are artificial intelligence models trained on vast text data to generate humanlike outputs. They have been applied to various tasks in health care, ranging from answering medical examination questions to generating clinical reports. With increasing institutional partnerships between companies producing LLMs and health systems, the real-world clinical application of these models is nearing realization. As these models gain traction, health care practitioners must understand what LLMs are, their development, their current and potential applications, and the associated pitfalls in a medical setting. This review, coupled with a tutorial, provides a comprehensive yet accessible overview of these areas with the aim of familiarizing health care professionals with the rapidly changing landscape of LLMs in medicine. Furthermore, the authors highlight active research areas in the field that promise to improve LLMs' usability in health care contexts.

Black Box Warning: Large Language Models and the Future of Infectious Diseases Consultation.

Large language models (LLMs) are artificial intelligence systems trained by deep learning algorithms to process natural language and generate text responses to user prompts. Some approach physician performance on a range of medical challenges, leading some proponents to advocate for their potential use in clinical consultation and prompting some consternation about the future of cognitive specialties. However, LLMs currently have limitations that preclude safe clinical deployment in performing specialist consultations, including frequent confabulations, lack of contextual awareness crucial for nuanced diagnostic and treatment plans, inscrutable and unexplainable training data and methods, and propensity to recapitulate biases. Nonetheless, considering the rapid improvement in this technology, growing calls for clinical integration, and healthcare systems that chronically undervalue cognitive specialties, it is critical that infectious diseases clinicians engage with LLMs to enable informed advocacy for how they should-and shouldn't-be used to augment specialist care.

Pharmacogenomics and big genomic data: from lab to clinic and back again.

The field of pharmacogenomics is an area of great potential for near-term human health impacts from the big genomic data revolution. Pharmacogenomics research momentum is building with numerous hypotheses currently being investigated through the integration of molecular profiles of different cell lines and large genomic data sets containing information on cellular and human responses to therapies. Additionally, the results of previous pharmacogenetic research efforts have been formulated into clinical guidelines that are beginning to impact how healthcare is conducted on the level of the individual patient. This trend will only continue with the recent release of new datasets containing linked genotype and electronic medical record data. This review discusses key resources available for pharmacogenomics and pharmacogenetics research and highlights recent work within the field.

Predicting venous thromboembolism risk from exomes in the Critical Assessment of Genome Interpretation (CAGI) challenges.

Genetics play a key role in venous thromboembolism (VTE) risk, however established risk factors in European populations do not translate to individuals of African descent because of the differences in allele frequencies between populations. As part of the fifth iteration of the Critical Assessment of Genome Interpretation, participants were asked to predict VTE status in exome data from African American subjects. Participants were provided with 103 unlabeled exomes from patients treated with warfarin for non-VTE causes or VTE and asked to predict which disease each subject had been treated for. Given the lack of training data, many participants opted to use unsupervised machine learning methods, clustering the exomes by variation in genes known to be associated with VTE. The best performing method using only VTE related genes achieved an area under the ROC curve of 0.65. Here, we discuss the range of methods used in the prediction of VTE from sequence data and explore some of the difficulties of conducting a challenge with known confounders. In addition, we show that an existing genetic risk score for VTE that was developed in European subjects works well in African Americans.

Working toward precision medicine: Predicting phenotypes from exomes in the Critical Assessment of Genome Interpretation (CAGI) challenges.

Precision medicine aims to predict a patient's disease risk and best therapeutic options by using that individual's genetic sequencing data. The Critical Assessment of Genome Interpretation (CAGI) is a community experiment consisting of genotype-phenotype prediction challenges; participants build models, undergo assessment, and share key findings. For CAGI 4, three challenges involved using exome-sequencing data: Crohn's disease, bipolar disorder, and warfarin dosing. Previous CAGI challenges included prior versions of the Crohn's disease challenge. Here, we discuss the range of techniques used for phenotype prediction as well as the methods used for assessing predictive models. Additionally, we outline some of the difficulties associated with making predictions and evaluating them. The lessons learned from the exome challenges can be applied to both research and clinical efforts to improve phenotype prediction from genotype. In addition, these challenges serve as a vehicle for sharing clinical and research exome data in a secure manner with scientists who have a broad range of expertise, contributing to a collaborative effort to advance our understanding of genotype-phenotype relationships.

Targeted exon capture and sequencing in sporadic amyotrophic lateral sclerosis.

Amyotrophic lateral sclerosis (ALS) is a devastating neurodegenerative disease that results in progressive degeneration of motor neurons, ultimately leading to paralysis and death. Approximately 10% of ALS cases are familial, with the remaining 90% of cases being sporadic. Genetic studies in familial cases of ALS have been extremely informative in determining the causative mutations behind ALS, especially as the same mutations identified in familial ALS can also cause sporadic disease. However, the cause of ALS in approximately 30% of familial cases and in the majority of sporadic cases remains unknown. Sporadic ALS cases represent an underutilized resource for genetic information about ALS; therefore, we undertook a targeted sequencing approach of 169 known and candidate ALS disease genes in 242 sporadic ALS cases and 129 matched controls to try to identify novel variants linked to ALS. We found a significant enrichment in novel and rare variants in cases versus controls, indicating that we are likely identifying disease associated mutations. This study highlights the utility of next generation sequencing techniques combined with functional studies and rare variant analysis tools to provide insight into the genetic etiology of a heterogeneous sporadic disease.

Genetic variant in folate homeostasis is associated with lower warfarin dose in African Americans.

The anticoagulant warfarin has >30 million prescriptions per year in the United States. Doses can vary 20-fold between patients, and incorrect dosing can result in serious adverse events. Variation in warfarin pharmacokinetic and pharmacodynamic genes, such as CYP2C9 and VKORC1, do not fully explain the dose variability in African Americans. To identify additional genetic contributors to warfarin dose, we exome sequenced 103 African Americans on stable doses of warfarin at extremes (≤ 35 and ≥ 49 mg/week). We found an association between lower warfarin dose and a population-specific regulatory variant, rs7856096 (P = 1.82 × 10(-8), minor allele frequency = 20.4%), in the folate homeostasis gene folylpolyglutamate synthase (FPGS). We replicated this association in an independent cohort of 372 African American subjects whose stable warfarin doses represented the full dosing spectrum (P = .046). In a combined cohort, adding rs7856096 to the International Warfarin Pharmacogenetic Consortium pharmacogenetic dosing algorithm resulted in a 5.8 mg/week (P = 3.93 × 10(-5)) decrease in warfarin dose for each allele carried. The variant overlaps functional elements and was associated (P = .01) with FPGS gene expression in lymphoblastoid cell lines derived from combined HapMap African populations (N = 326). Our results provide the first evidence linking genetic variation in folate homeostasis to warfarin response.

Genetic variants associated with warfarin dose in African-American individuals: a genome-wide association study.

BACKGROUND: VKORC1 and CYP2C9 are important contributors to warfarin dose variability, but explain less variability for individuals of African descent than for those of European or Asian descent. We aimed to identify additional variants contributing to warfarin dose requirements in African Americans. METHODS: We did a genome-wide association study of discovery and replication cohorts. Samples from African-American adults (aged ≥18 years) who were taking a stable maintenance dose of warfarin were obtained at International Warfarin Pharmacogenetics Consortium (IWPC) sites and the University of Alabama at Birmingham (Birmingham, AL, USA). Patients enrolled at IWPC sites but who were not used for discovery made up the independent replication cohort. All participants were genotyped. We did a stepwise conditional analysis, conditioning first for VKORC1 -1639G→A, followed by the composite genotype of CYP2C9*2 and CYP2C9*3. We prespecified a genome-wide significance threshold of p<5×10(-8) in the discovery cohort and p<0·0038 in the replication cohort. FINDINGS: The discovery cohort contained 533 participants and the replication cohort 432 participants. After the prespecified conditioning in the discovery cohort, we identified an association between a novel single nucleotide polymorphism in the CYP2C cluster on chromosome 10 (rs12777823) and warfarin dose requirement that reached genome-wide significance (p=1·51×10(-8)). This association was confirmed in the replication cohort (p=5·04×10(-5)); analysis of the two cohorts together produced a p value of 4·5×10(-12). Individuals heterozygous for the rs12777823 A allele need a dose reduction of 6·92 mg/week and those homozygous 9·34 mg/week. Regression analysis showed that the inclusion of rs12777823 significantly improves warfarin dose variability explained by the IWPC dosing algorithm (21% relative improvement). INTERPRETATION: A novel CYP2C single nucleotide polymorphism exerts a clinically relevant effect on warfarin dose in African Americans, independent of CYP2C9*2 and CYP2C9*3. Incorporation of this variant into pharmacogenetic dosing algorithms could improve warfarin dose prediction in this population. FUNDING: National Institutes of Health, American Heart Association, Howard Hughes Medical Institute, Wisconsin Network for Health Research, and the Wellcome Trust.

The Promises and Perils of Foundation Models in Dermatology.

Foundation models (FM), which are large-scale artificial intelligence (AI) models that can complete a range of tasks, represent a paradigm shift in AI. These versatile models encompass large language models, vision-language models, and multimodal models. Although these models are often trained for broad tasks, they have been applied either out of the box or after additional fine tuning to tasks in medicine, including dermatology. From addressing administrative tasks to answering dermatology questions, these models are poised to have an impact on dermatology care delivery. As FMs become more ubiquitous in health care, it is important for clinicians and dermatologists to have a basic understanding of how these models are developed, what they are capable of, and what pitfalls exist. In this paper, we present a comprehensive yet accessible overview of the current state of FMs and summarize their current applications in dermatology, highlight their limitations, and discuss future developments in the field.

Human-AI interaction in skin cancer diagnosis: a systematic review and meta-analysis.

The development of diagnostic tools for skin cancer based on artificial intelligence (AI) is increasing rapidly and will likely soon be widely implemented in clinical use. Even though the performance of these algorithms is promising in theory, there is limited evidence on the impact of AI assistance on human diagnostic decisions. Therefore, the aim of this systematic review and meta-analysis was to study the effect of AI assistance on the accuracy of skin cancer diagnosis. We searched PubMed, Embase, IEE Xplore, Scopus and conference proceedings for articles from 1/1/2017 to 11/8/2022. We included studies comparing the performance of clinicians diagnosing at least one skin cancer with and without deep learning-based AI assistance. Summary estimates of sensitivity and specificity of diagnostic accuracy with versus without AI assistance were computed using a bivariate random effects model. We identified 2983 studies, of which ten were eligible for meta-analysis. For clinicians without AI assistance, pooled sensitivity was 74.8% (95% CI 68.6-80.1) and specificity was 81.5% (95% CI 73.9-87.3). For AI-assisted clinicians, the overall sensitivity was 81.1% (95% CI 74.4-86.5) and specificity was 86.1% (95% CI 79.2-90.9). AI benefitted medical professionals of all experience levels in subgroup analyses, with the largest improvement among non-dermatologists. No publication bias was detected, and sensitivity analysis revealed that the findings were robust. AI in the hands of clinicians has the potential to improve diagnostic accuracy in skin cancer diagnosis. Given that most studies were conducted in experimental settings, we encourage future studies to further investigate these potential benefits in real-life settings.

Current State of Dermatology Mobile Applications With Artificial Intelligence Features.

Importance: With advancements in mobile technology and artificial intelligence (AI) methods, there has been a substantial surge in the availability of direct-to-consumer mobile applications (apps) claiming to aid in the assessment and management of diverse skin conditions. Despite widespread patient downloads, these apps exhibit limited evidence supporting their efficacy. Objective: To identify and characterize current English-language AI dermatology mobile apps available for download, focusing on aspects such as purpose, supporting evidence, regulatory status, clinician input, data privacy measures, and use of image data. Evidence Review: In this cross-sectional study, both Apple and Android mobile app stores were systematically searched for dermatology-related apps that use AI algorithms. Each app's purpose, target audience, evidence-based claims, algorithm details, data availability, clinician input during development, and data usage privacy policies were evaluated. Findings: A total of 909 apps were initially identified. Following the removal of 518 duplicates, 391 apps remained. Subsequent review excluded 350 apps due to nonmedical nature, non-English languages, absence of AI features, or unavailability, ultimately leaving 41 apps for detailed analysis. The findings revealed several concerning aspects of the current landscape of AI apps in dermatology. Notably, none of the apps were approved by the US Food and Drug Administration, and only 2 of the apps included disclaimers for the lack of regulatory approval. Overall, the study found that these apps lack supporting evidence, input from clinicians and/or dermatologists, and transparency in algorithm development, data usage, and user privacy. Conclusions and Relevance: This cross-sectional study determined that although AI dermatology mobile apps hold promise for improving access to care and patient outcomes, in their current state, they may pose harm due to potential risks, lack of consistent validation, and misleading user communication. Addressing challenges in efficacy, safety, and transparency through effective regulation, validation, and standardized evaluation criteria is essential to harness the benefits of these apps while minimizing risks.

Session Introduction: Artificial Intelligence in Clinical Medicine: Generative and Interactive Systems at the Human-Machine Interface.

Artificial Intelligence (AI) models are substantially enhancing the capability to analyze complex and multi-dimensional datasets. Generative AI and deep learning models have demonstrated significant advancements in extracting knowledge from unstructured text, imaging as well as structured and tabular data. This recent breakthrough in AI has inspired research in medicine, leading to the development of numerous tools for creating clinical decision support systems, monitoring tools, image interpretation, and triaging capabilities. Nevertheless, comprehensive research is imperative to evaluate the potential impact and implications of AI systems in healthcare. At the 2024 Pacific Symposium on Biocomputing (PSB) session entitled "Artificial Intelligence in Clinical Medicine: Generative and Interactive Systems at the Human-Machine Interface", we spotlight research that develops and applies AI algorithms to solve real-world problems in healthcare.

Transparent medical image AI via an image-text foundation model grounded in medical literature.

Building trustworthy and transparent image-based medical artificial intelligence (AI) systems requires the ability to interrogate data and models at all stages of the development pipeline, from training models to post-deployment monitoring. Ideally, the data and associated AI systems could be described using terms already familiar to physicians, but this requires medical datasets densely annotated with semantically meaningful concepts. In the present study, we present a foundation model approach, named MONET (medical concept retriever), which learns how to connect medical images with text and densely scores images on concept presence to enable important tasks in medical AI development and deployment such as data auditing, model auditing and model interpretation. Dermatology provides a demanding use case for the versatility of MONET, due to the heterogeneity in diseases, skin tones and imaging modalities. We trained MONET based on 105,550 dermatological images paired with natural language descriptions from a large collection of medical literature. MONET can accurately annotate concepts across dermatology images as verified by board-certified dermatologists, competitively with supervised models built on previously concept-annotated dermatology datasets of clinical images. We demonstrate how MONET enables AI transparency across the entire AI system development pipeline, from building inherently interpretable models to dataset and model auditing, including a case study dissecting the results of an AI clinical trial.

Single-Lesion Skin Cancer Risk Stratification Triage Pathway.

Importance: Access to timely dermatologic care remains a challenge, especially for patients with new skin lesions. Assessing the efficiency of new triage pathways may assist in better resource allocation and shorter time to care. Objective: To evaluate whether a rule-based triage system was associated with better skin cancer risk stratification of patients and reduced wait times. Design, Setting, and Participants: This retrospective quality improvement cohort study of patients referred to Stanford University dermatology clinics was conducted between November 2017 and January 2023. A rules-based triage system based on a priori-determined high-risk lesion characteristics was implemented. Exposures: Referral reasons and risk factors of patients provided by their primary care physicians. Main Outcomes and Measures: Biopsy results of patients (diagnosis of any skin cancer and melanoma) at their visit or within 6 months after the visit. Regression models were used to assess the association between risk factors at referral and (1) biopsy outcomes and (2) time to first visit, adjusting for sociodemographic factors. Results: Among 37 478 patients (mean [SD] age, 54 (18) years; 21 292 women [57%]), the rates of aggregate biopsy, malignant biopsy specimens, and melanoma were comparable across patients seen after (n = 12 302) and before (n = 25 176) the implementation of the new triage pathway. Patients seen through the lesion pathway had a higher risk of having malignant biopsy results (adjusted risk ratio [aRR], 1.6; 95% CI, 1.4-1.9) and melanoma (aRR, 2.0; 95% CI, 1.2-3.2) than those not seen through the pathway. Lesions that were concerning to referring clinicians for skin cancer were associated with an increased risk of skin cancer (all skin cancer: aRR, 2.8; 95% CI, 2.2-3.5; melanoma: aRR, 2.02; 95% CI, 1.1-3.7). Patients in the 3 high-risk lesion groups were seen faster in the new triage pathway (mean reduction, 26 days; 95% CI, 18-34 days). Conclusions and Relevance: In this study, a new automated, rules-based referral pathway was implemented that expedited care for patients with high-risk skin cancer. This reform may have contributed to improving patient stratification, reducing the time from referral to first encounter, and maintaining accuracy in identifying malignant lesions. The findings highlight the potential to optimize clinical resource allocation by better risk stratification of referred patients.

Deep learning-aided decision support for diagnosis of skin disease across skin tones.

Although advances in deep learning systems for image-based medical diagnosis demonstrate their potential to augment clinical decision-making, the effectiveness of physician-machine partnerships remains an open question, in part because physicians and algorithms are both susceptible to systematic errors, especially for diagnosis of underrepresented populations. Here we present results from a large-scale digital experiment involving board-certified dermatologists (n = 389) and primary-care physicians (n = 459) from 39 countries to evaluate the accuracy of diagnoses submitted by physicians in a store-and-forward teledermatology simulation. In this experiment, physicians were presented with 364 images spanning 46 skin diseases and asked to submit up to four differential diagnoses. Specialists and generalists achieved diagnostic accuracies of 38% and 19%, respectively, but both specialists and generalists were four percentage points less accurate for the diagnosis of images of dark skin as compared to light skin. Fair deep learning system decision support improved the diagnostic accuracy of both specialists and generalists by more than 33%, but exacerbated the gap in the diagnostic accuracy of generalists across skin tones. These results demonstrate that well-designed physician-machine partnerships can enhance the diagnostic accuracy of physicians, illustrating that success in improving overall diagnostic accuracy does not necessarily address bias.

Pathway analysis of genome-wide data improves warfarin dose prediction.

BACKGROUND: Many genome-wide association studies focus on associating single loci with target phenotypes. However, in the setting of rare variation, accumulating sufficient samples to assess these associations can be difficult. Moreover, multiple variations in a gene or a set of genes within a pathway may all contribute to the phenotype, suggesting that the aggregation of variations found over the gene or pathway may be useful for improving the power to detect associations. RESULTS: Here, we present a method for aggregating single nucleotide polymorphisms (SNPs) along biologically relevant pathways in order to seek genetic associations with phenotypes. Our method uses all available genetic variants and does not remove those in linkage disequilibrium (LD). Instead, it uses a novel SNP weighting scheme to down-weight the contributions of correlated SNPs. We apply our method to three cohorts of patients taking warfarin: two European descent cohorts and an African American cohort. Although the clinical covariates and key pharmacogenetic loci for warfarin have been characterized, our association metric identifies a significant association with mutations distributed throughout the pathway of warfarin metabolism. We improve dose prediction after using all known clinical covariates and pharmacogenetic variants in VKORC1 and CYP2C9. In particular, we find that at least 1% of the missing heritability in warfarin dose may be due to the aggregated effects of variations in the warfarin metabolic pathway, even though the SNPs do not individually show a significant association. CONCLUSIONS: Our method allows researchers to study aggregative SNP effects in an unbiased manner by not preselecting SNPs. It retains all the available information by accounting for LD-structure through weighting, which eliminates the need for LD pruning.