Rajat Vashistha, Arif Gulzar, Parveen Kundu
et al.
India faces a significant cancer burden, with an incidence-to-mortality ratio indicating that nearly three out of five individuals diagnosed with cancer succumb to the disease. While the limitations of physical healthcare infrastructure are widely acknowledged as a primary challenge, concerted efforts by government and healthcare agencies are underway to mitigate these constraints. However, given the country's vast geography and high population density, it is imperative to explore alternative soft infrastructure solutions to complement existing frameworks. Artificial Intelligence agents are increasingly transforming problem-solving approaches across various domains, with their application in medicine proving particularly transformative. In this perspective, we examine the potential role of AI agents in advancing nuclear medicine for cancer research, diagnosis, and management in India. We begin with a brief overview of AI agents and their capabilities, followed by a proposed agent-based ecosystem that can address prevailing sustainability challenges in India nuclear medicine.
This paper explores the application of prompt engineering to enhance the performance of large language models (LLMs) in the domain of Traditional Chinese Medicine (TCM). We propose TCM-Prompt, a framework that integrates various pre-trained language models (PLMs), templates, tokenization, and verbalization methods, allowing researchers to easily construct and fine-tune models for specific TCM-related tasks. We conducted experiments on disease classification, syndrome identification, herbal medicine recommendation, and general NLP tasks, demonstrating the effectiveness and superiority of our approach compared to baseline methods. Our findings suggest that prompt engineering is a promising technique for improving the performance of LLMs in specialized domains like TCM, with potential applications in digitalization, modernization, and personalized medicine.
Technologies adopted by the public sector have transformed the work practices of employees in public agencies by creating different means of communication and decision-making. Although much of the recent research in the future of work domain has concentrated on the effects of technological advancements on public sector employees, the influence on work practices of external stakeholders engaging with this sector remains under-explored. In this paper, we focus on a digital platform called OneStop which is deployed by several building departments across the U.S. and aims to integrate various steps and services into a single point of online contact between public sector employees and the public. Drawing on semi-structured interviews with 22 stakeholders, including local business owners, experts involved in the construction process, community representatives, and building department employees, we investigate how this technology transition has impacted the work of these different stakeholders. We observe a multifaceted perspective and experience caused by the adoption of OneStop. OneStop exacerbated inequitable practices for local business owners due to a lack of face-to-face interactions with the department employees. For the public sector employees, OneStop standardized the work practices, representing the building department's priorities and values. Based on our findings, we discuss tensions around standardization, equality, and equity in technology transition, as well as design implications for equitable practices in the public sector.
Since the release of ChatGPT and GPT-4, large language models (LLMs) and multimodal large language models (MLLMs) have attracted widespread attention for their exceptional capabilities in understanding, reasoning, and generation, introducing transformative paradigms for integrating artificial intelligence into medicine. This survey provides a comprehensive overview of the development, principles, application scenarios, challenges, and future directions of LLMs and MLLMs in medicine. Specifically, it begins by examining the paradigm shift, tracing the transition from traditional models to LLMs and MLLMs, and highlighting the unique advantages of these LLMs and MLLMs in medical applications. Next, the survey reviews existing medical LLMs and MLLMs, providing detailed guidance on their construction and evaluation in a clear and systematic manner. Subsequently, to underscore the substantial value of LLMs and MLLMs in healthcare, the survey explores five promising applications in the field. Finally, the survey addresses the challenges confronting medical LLMs and MLLMs and proposes practical strategies and future directions for their integration into medicine. In summary, this survey offers a comprehensive analysis of the technical methodologies and practical clinical applications of medical LLMs and MLLMs, with the goal of bridging the gap between these advanced technologies and clinical practice, thereby fostering the evolution of the next generation of intelligent healthcare systems.
As Artificial Intelligence (AI) increasingly influences various aspects of society, there is growing public interest in its potential benefits and risks. In this paper we present results of public perception of AI from a survey conducted with 10,000 respondents spanning ten countries in four continents around the world. The results show that currently an equal percentage of respondents who believe AI will change the world as we know it, also believe AI needs to be heavily regulated. However, our findings also indicate that despite the general sentiment among the global public that AI will replace workers, if a company were to use AI to innovate to improve lives, the public would be more likely to think highly of the company, purchase from them and even be interested in a job in that company. Our results further reveal that the global public largely views AI as a tool for problem solving. These nuanced results underscore the importance of AI directed towards challenges that the public would like science and technology-based innovations to address. We draw on a multi-year 3M study of public perception of science to provide further context on what the public perceives as important problems to be solved.
Excellence in a wide variety of medical applications poses considerable challenges for AI, requiring advanced reasoning, access to up-to-date medical knowledge and understanding of complex multimodal data. Gemini models, with strong general capabilities in multimodal and long-context reasoning, offer exciting possibilities in medicine. Building on these core strengths of Gemini, we introduce Med-Gemini, a family of highly capable multimodal models that are specialized in medicine with the ability to seamlessly use web search, and that can be efficiently tailored to novel modalities using custom encoders. We evaluate Med-Gemini on 14 medical benchmarks, establishing new state-of-the-art (SoTA) performance on 10 of them, and surpass the GPT-4 model family on every benchmark where a direct comparison is viable, often by a wide margin. On the popular MedQA (USMLE) benchmark, our best-performing Med-Gemini model achieves SoTA performance of 91.1% accuracy, using a novel uncertainty-guided search strategy. On 7 multimodal benchmarks including NEJM Image Challenges and MMMU (health & medicine), Med-Gemini improves over GPT-4V by an average relative margin of 44.5%. We demonstrate the effectiveness of Med-Gemini's long-context capabilities through SoTA performance on a needle-in-a-haystack retrieval task from long de-identified health records and medical video question answering, surpassing prior bespoke methods using only in-context learning. Finally, Med-Gemini's performance suggests real-world utility by surpassing human experts on tasks such as medical text summarization, alongside demonstrations of promising potential for multimodal medical dialogue, medical research and education. Taken together, our results offer compelling evidence for Med-Gemini's potential, although further rigorous evaluation will be crucial before real-world deployment in this safety-critical domain.
Extracting medication names from handwritten doctor prescriptions is challenging due to the wide variability in handwriting styles and prescription formats. This paper presents a robust method for extracting medicine names using a combination of Mask R-CNN and Transformer-based Optical Character Recognition (TrOCR) with Multi-Head Attention and Positional Embeddings. A novel dataset, featuring diverse handwritten prescriptions from various regions of Pakistan, was utilized to fine-tune the model on different handwriting styles. The Mask R-CNN model segments the prescription images to focus on the medicinal sections, while the TrOCR model, enhanced by Multi-Head Attention and Positional Embeddings, transcribes the isolated text. The transcribed text is then matched against a pre-existing database for accurate identification. The proposed approach achieved a character error rate (CER) of 1.4% on standard benchmarks, highlighting its potential as a reliable and efficient tool for automating medicine name extraction.
Yoyok Bekti Prasetyo, Faridi Faridi, Nur Lailatul Masruroh
et al.
Introduction: Family resilience is strongly influenced by religious coping and spiritual/religious well-being (RW). In the context of the COVID-19 pandemic in Indonesia, this study intends to investigate the relationship between religious coping, spiritual well-being, and family resilience. Methods: A cross-sectional survey (n = 242) was conducted from December 2021 to January 2022 in Indonesia. The Spiritual Coping Strategies Scale-Chinese version, Spiritual Well-Being Scale, and Family Resilience Assessment Scale were used for data collection. Smart Partial Least Square (SmartPLS) software (version 3.2.7) was used to analyze the data. Results: Most respondents aged range from 46 to 55 years-old (30.9%). Also, most of them were having senior high school educational level (47.7%), earn <3 million rupiah (90.5%), and jobless (66.7%). Family resilience to COVID-19 has been influenced by the relationship between RW and existential well-being (EW) (81.2%) (β =0.901, t = 24,836, P = 0.001). Religious Non-coping (RNC)- Religious well-being(RW) is 0.124, which indicating that RNC affecting RW by 12.4%, and it also impacting on family resilience to COVID-19 (β = −0.310, t = 3.275, P = 0.001, f2 = 0.085; minor). Conclusion: Religious coping, RW, and EW are all important factors influencing family resilience during the COVID-19 pandemic. Currently, the COVID-19 pandemic has ended. However, with the dynamic development of world health, an outbreak may occur in the future, so the findings of this research will be helpful in providing a warning about spiritual factors that significantly influence family resilience.
Large language models (LLMs), such as ChatGPT, have received substantial attention due to their capabilities for understanding and generating human language. While there has been a burgeoning trend in research focusing on the employment of LLMs in supporting different medical tasks (e.g., enhancing clinical diagnostics and providing medical education), a review of these efforts, particularly their development, practical applications, and outcomes in medicine, remains scarce. Therefore, this review aims to provide a detailed overview of the development and deployment of LLMs in medicine, including the challenges and opportunities they face. In terms of development, we provide a detailed introduction to the principles of existing medical LLMs, including their basic model structures, number of parameters, and sources and scales of data used for model development. It serves as a guide for practitioners in developing medical LLMs tailored to their specific needs. In terms of deployment, we offer a comparison of the performance of different LLMs across various medical tasks, and further compare them with state-of-the-art lightweight models, aiming to provide an understanding of the advantages and limitations of LLMs in medicine. Overall, in this review, we address the following questions: 1) What are the practices for developing medical LLMs 2) How to measure the medical task performance of LLMs in a medical setting? 3) How have medical LLMs been employed in real-world practice? 4) What challenges arise from the use of medical LLMs? and 5) How to more effectively develop and deploy medical LLMs? By answering these questions, this review aims to provide insights into the opportunities for LLMs in medicine and serve as a practical resource. We also maintain a regularly updated list of practical guides on medical LLMs at https://github.com/AI-in-Health/MedLLMsPracticalGuide
Extensive adoption of electronic health records (EHRs) offers opportunities for their use in various downstream clinical analyses. To accomplish this purpose, enriching an EHR cohort with external knowledge (e.g., standardized medical ontology and wealthy semantics) could help us reveal more comprehensive insights via a spectrum of informative relations among medical codes. Nevertheless, harnessing those beneficial interconnections was scarcely exercised, especially in the medication recommendation task. This study proposes a novel Knowledge-Induced Medicine Prescribing Network (KindMed) to recommend medicines by inducing knowledge from myriad medical-related external sources upon the EHR cohort and rendering interconnected medical codes as medical knowledge graphs (KGs). On top of relation-aware graph representation learning to obtain an adequate embedding over such KGs, we leverage hierarchical sequence learning to discover and fuse temporal dynamics of clinical (i.e., diagnosis and procedures) and medicine streams across patients' historical admissions to foster personalized recommendations. Eventually, we employ attentive prescribing that accounts for three essential patient representations, i.e., a summary of joint historical medical records, clinical progression, and the current clinical state of patients. We validated the effectiveness of our KindMed on the augmented real-world EHR cohorts, achieving improved recommendation performances against a handful of graph-driven baselines.
Haider Sultan, Hafiza Farwa Mahmood, Noor Fatima
et al.
Like other fields of Traditional Medicines, Unani Medicines have been found as an effective medical practice for ages. It is still widely used in the subcontinent, particularly in Pakistan and India. However, Unani Medicines Practitioners are lacking modern IT applications in their everyday clinical practices. An Online Clinical Decision Support System may address this challenge to assist apprentice Unani Medicines practitioners in their diagnostic processes. The proposed system provides a web-based interface to enter the patient's symptoms, which are then automatically analyzed by our system to generate a list of probable diseases. The system allows practitioners to choose the most likely disease and inform patients about the associated treatment options remotely. The system consists of three modules: an Online Clinical Decision Support System, an Artificial Intelligence Inference Engine, and a comprehensive Unani Medicines Database. The system employs advanced AI techniques such as Decision Trees, Deep Learning, and Natural Language Processing. For system development, the project team used a technology stack that includes React, FastAPI, and MySQL. Data and functionality of the application is exposed using APIs for integration and extension with similar domain applications. The novelty of the project is that it addresses the challenge of diagnosing diseases accurately and efficiently in the context of Unani Medicines principles. By leveraging the power of technology, the proposed Clinical Decision Support System has the potential to ease access to healthcare services and information, reduce cost, boost practitioner and patient satisfaction, improve speed and accuracy of the diagnostic process, and provide effective treatments remotely. The application will be useful for Unani Medicines Practitioners, Patients, Government Drug Regulators, Software Developers, and Medical Researchers.
Murali Shunmugam, Venkatesan Chakrapani, Pushpesh Kumar
et al.
Background: Smartphone use is increasing among transgender women (TGW), including those who engage in sex work. Current government-supported HIV prevention interventions focus on physical venue-based outreach among TGW, missing the opportunity to reach them through smartphone-based interventions. Objective: We examined the use of smartphones among TGW, especially in relation to social and sexual networking, and explored their perspectives on their willingness to use smartphone-based HIV prevention interventions. Materials and Methods: Through an exploratory descriptive-interpretive qualitative research design, we conducted 6 focus groups with a purposive sample of 30 TGW (70% in sex work) and 4 key informant in-depth interviews in Chennai and Hyderabad, India. Data were explored using framework analysis. Results: Through smartphones, TGW used social media (e.g., WhatsApp and Facebook) and dating applications for socialization, meeting sexual partners, and entertainment. Low-literacy TGW used voice or video messaging. TGW expressed interest in receiving short health-related videos and text messages on HIV, mental health, and gender transition. Conclusion: At-risk TGW could potentially be reached through smartphone-based online health promotion interventions, but those interventions need to be holistic – moving beyond HIV.
With the increasing popularity of graph neural networks (GNNs) in several sensitive applications like healthcare and medicine, concerns have been raised over the privacy aspects of trained GNNs. More notably, GNNs are vulnerable to privacy attacks, such as membership inference attacks, even if only black-box access to the trained model is granted. We propose PrivGNN, a privacy-preserving framework for releasing GNN models in a centralized setting. Assuming an access to a public unlabeled graph, PrivGNN provides a framework to release GNN models trained explicitly on public data along with knowledge obtained from the private data in a privacy preserving manner. PrivGNN combines the knowledge-distillation framework with the two noise mechanisms, random subsampling, and noisy labeling, to ensure rigorous privacy guarantees. We theoretically analyze our approach in the Renyi differential privacy framework. Besides, we show the solid experimental performance of our method compared to several baselines adapted for graph-structured data. Our code is available at https://github.com/iyempissy/privGnn.
The increasing prevalence and worldwide distribution of multidrug-resistant bacterial pathogens is an imminent danger to public health and threatens virtually all aspects of modern medicine. Particularly concerning, yet insufficiently addressed, are the members of the Burkholderia cepacia complex (Bcc), a group of at least twenty opportunistic, hospital-transmitted, and notoriously drug-resistant species, which infect and cause morbidity in patients who are immunocompromised and those afflicted with chronic illnesses, including cystic fibrosis (CF) and chronic granulomatous disease (CGD). One potential solution to the antimicrobial resistance crisis is phage therapy—the use of phages for the treatment of bacterial infections. Although phage therapy has a long and somewhat checkered history, an impressive volume of modern research has been amassed in the past decades to show that when applied through specific, scientifically supported treatment strategies, phage therapy is highly efficacious and is a promising avenue against drug-resistant and difficult-to-treat pathogens, such as the Bcc. In this review, we discuss the clinical significance of the Bcc, the advantages of phage therapy, and the theoretical and clinical advancements made in phage therapy in general over the past decades, and apply these concepts specifically to the nascent, but growing and rapidly developing, field of Bcc phage therapy.
The coronavirus disease 2019 (COVID-19) pandemic has caused major disruptions in all aspects of daily life, from school and work to interactions with friends and families. Mitigation measures also substantially altered the economic environment, with tens of millions of people in the US losing their jobs, and many more experiencing income reductions (through furloughs) or uncertainty about the future of employment and health insurance. In addition, major changes took place almost overnight in the landscape of medicine and medical care, including new policies to reduce social/physical interaction and cessation of many in-person medical visits. Within the field of psychiatry, a rapidly growing literature is addressing the disruption and transformation experienced during the pandemic; more than 1000 publications have already appeared. This Viewpoint describes some of these developments to date and discusses important themes relevant to clinical psychiatry, care delivery challenges, and public health considerations.
I. A. Kirshchina, T. V. Shestakova, A. V. Soloninina
et al.
Preservation of public health is the main goal of social progress and development of the society. The search for potential opportunities to improve individual and public health indicators is a positive predictor of increasing the socio-economic efficiency of the society and increasing the healthy life expectancy of citizens. The scientific review provides arguments in favor of the need for professional collaboration of specialists from various industries in order to universally realize the most important right of citizens to protect their own and public health. Regulatory legal documents defining national and international policy in the field of health protection and forming the general vector of development of health care activities were used as the sources of information for the formation of the basis of the study. As a result of a logical generalization of global and national priorities and trends in the development of the healthcare sector, the main prerequisites (determinants) of health saving of citizens, adopted by the world community and reflected in domestic documents, are formulated. The main prerequisites (determinants) of health are defined as: promotion of activities that advantage health protection; creation of a single preventive space; specification of the concept of 'responsible attitude to health'; coverage of the entire life cycle of a person and all spheres of his activity in the formation of a responsible attitude to health and motivation for its preservation; development of information technologies in the field of health protection; expansion of intersectoral and interdisciplinary cooperation in order to maintain and strengthen health; improvement of public health literacy; transformation of health services from the standpoint of health protection; development of human resources to ensure health-saving activities. The identified determinants of the preservation of individual and public health can act as a theoretical basis for the development of a scientific and practical methodology aimed at solving problems of improving health through the potential of interdisciplinary interaction of specialists in various fields of activity.
Network Medicine has improved the mechanistic understanding of disease, offering quantitative insights into disease mechanisms, comorbidities, and novel diagnostic tools and therapeutic treatments. Yet, most network-based approaches rely on a comprehensive map of protein-protein interactions, ignoring interactions mediated by non-coding RNAs (ncRNAs). Here, we systematically combine experimentally confirmed binding interactions mediated by ncRNA with protein-protein interactions, constructing the first comprehensive network of all physical interactions in the human cell. We find that the inclusion of ncRNA, expands the number of genes in the interactome by 46% and the number of interactions by 107%, significantly enhancing our ability to identify disease modules. Indeed, we find that 132 diseases, lacked a statistically significant disease module in the protein-based interactome, but have a statistically significant disease module after inclusion of ncRNA-mediated interactions, making these diseases accessible to the tools of network medicine. We show that the inclusion of ncRNAs helps unveil disease-disease relationships that were not detectable before and expands our ability to predict comorbidity patterns between diseases. Taken together, we find that including non-coding interactions improves both the breath and the predictive accuracy of network medicine.
Precision medicine provides customized treatments to patients based on their characteristics and is a promising approach to improving treatment efficiency. Large scale omics data are useful for patient characterization, but often their measurements change over time, leading to longitudinal data. Random forest is one of the state-of-the-art machine learning methods for building prediction models, and can play a crucial role in precision medicine. In this paper, we review extensions of the standard random forest method for the purpose of longitudinal data analysis. Extension methods are categorized according to the data structures for which they are designed. We consider both univariate and multivariate responses and further categorize the repeated measurements according to whether the time effect is relevant. Information of available software implementations of the reviewed extensions is also given. We conclude with discussions on the limitations of our review and some future research directions.