Hasil untuk "Computer applications to medicine. Medical informatics"

F. Grajales, S. Sheps, K. Ho et al.

Background Social media are dynamic and interactive computer-mediated communication tools that have high penetration rates in the general population in high-income and middle-income countries. However, in medicine and health care, a large number of stakeholders (eg, clinicians, administrators, professional colleges, academic institutions, ministries of health, among others) are unaware of social media’s relevance, potential applications in their day-to-day activities, as well as the inherent risks and how these may be attenuated and mitigated. Objective We conducted a narrative review with the aim to present case studies that illustrate how, where, and why social media are being used in the medical and health care sectors. Methods Using a critical-interpretivist framework, we used qualitative methods to synthesize the impact and illustrate, explain, and provide contextual knowledge of the applications and potential implementations of social media in medicine and health care. Both traditional (eg, peer-reviewed) and nontraditional (eg, policies, case studies, and social media content) sources were used, in addition to an environmental scan (using Google and Bing Web searches) of resources. Results We reviewed, evaluated, and synthesized 76 articles, 44 websites, and 11 policies/reports. Results and case studies are presented according to 10 different categories of social media: (1) blogs (eg, WordPress), (2) microblogs (eg, Twitter), (3) social networking sites (eg, Facebook), (4) professional networking sites (eg, LinkedIn, Sermo), (5) thematic networking sites (eg, 23andMe), (6) wikis (eg, Wikipedia), (7) mashups (eg, HealthMap), (8) collaborative filtering sites (eg, Digg), (9) media sharing sites (eg, YouTube, Slideshare), and others (eg, SecondLife). Four recommendations are provided and explained for stakeholders wishing to engage with social media while attenuating risk: (1) maintain professionalism at all times, (2) be authentic, have fun, and do not be afraid, (3) ask for help, and (4) focus, grab attention, and engage. Conclusions The role of social media in the medical and health care sectors is far reaching, and many questions in terms of governance, ethics, professionalism, privacy, confidentiality, and information quality remain unanswered. By following the guidelines presented, professionals have a starting point to engage with social media in a safe and ethical manner. Future research will be required to understand the synergies between social media and evidence-based practice, as well as develop institutional policies that benefit patients, clinicians, public health practitioners, and industry alike.

F. Kolbinger, G. P. Veldhuizen, Jiefu Zhu et al.

The field of Artificial Intelligence (AI) holds transformative potential in medicine. However, the lack of universal reporting guidelines poses challenges in ensuring the validity and reproducibility of published research studies in this field. Based on a systematic review of academic publications and reporting standards demanded by both international consortia and regulatory stakeholders as well as leading journals in the fields of medicine and medical informatics, 26 reporting guidelines published between 2009 and 2023 were included in this analysis. Guidelines were stratified by breadth (general or specific to medical fields), underlying consensus quality, and target research phase (preclinical, translational, clinical) and subsequently analyzed regarding the overlap and variations in guideline items. AI reporting guidelines for medical research vary with respect to the quality of the underlying consensus process, breadth, and target research phase. Some guideline items such as reporting of study design and model performance recur across guidelines, whereas other items are specific to particular fields and research stages. Our analysis highlights the importance of reporting guidelines in clinical AI research and underscores the need for common standards that address the identified variations and gaps in current guidelines. Overall, this comprehensive overview could help researchers and public stakeholders reinforce quality standards for increased reliability, reproducibility, clinical validity, and public trust in AI research in healthcare. This could facilitate the safe, effective, and ethical translation of AI methods into clinical applications that will ultimately improve patient outcomes. Artificial Intelligence (AI) refers to computer systems that can perform tasks that normally require human intelligence, like recognizing patterns or making decisions. AI has the potential to transform healthcare, but research on AI in medicine needs clear rules so caregivers and patients can trust it. This study reviews and compares 26 existing guidelines for reporting on AI in medicine. The key differences between these guidelines are their target areas (medicine in general or specific medical fields), the ways they were created, and the research stages they address. While some key items like describing the AI model recurred across guidelines, others were specific to the research area. The analysis shows gaps and variations in current guidelines. Overall, transparent reporting is important, so AI research is reliable, reproducible, trustworthy, and safe for patients. This systematic review of guidelines aims to increase the transparency of AI research, supporting an ethical and safe progression of AI from research into clinical practice. Kolbinger, Veldhuizen et al. systematically review reporting guidelines for artificial intelligence (AI) methods in clinical research. They identify several essential, commonly recommended items on study design and model performance, while other items are specific to particular fields and research stages.

Shkelqim Sherifi

Recent advancements in computer vision, driven by artificial intelligence, have significantly enhanced monitoring systems. One notable application is traffic monitoring, which leverages computer vision alongside deep learning-based object detection and counting. We present an offline, real-time traffic monitoring system that couples a pre-trained YOLOv11 detector with BoT-SORT/ByteTrack for multi-object tracking, implemented in PyTorch/OpenCV and wrapped in a Qt-based desktop UI. The CNN pipeline enables efficient vehicle detection and counting from video streams without cloud dependencies. Across diverse scenes, the system achieves (66.67-95.83%) counting accuracy. Class-wise detection yields high precision (cars: 0.97-1.00; trucks: 1.00) with strong recall (cars: 0.82-1.00; trucks: 0.70-1.00), resulting in F1 scores of (0.90-1.00 for cars and 0.82-1.00 for trucks). While adverse weather conditions may negatively impact this performance, results remain robust in typical conditions. By integrating lightweight models with an accessible, cloud-independent interface, this paper contributes to the modernization and development of future smart cities by showing the capacity of AI-driven traffic monitoring systems.

Liliana Hidalgo-Padilla, Massar Dabbous, Kristoffer Halvorsrud et al.

AbstractThe COVID-19 pandemic accelerated the adoption of remote consultations across health care, requiring rapid adjustments in service delivery. Consequently, there is an urgent need to understand the impact of remote consultations on health pathways. This viewpoint paper explores key challenges in data sources in England that hinder research on the impact of remote consultations on health outcomes. Based on our experience conducting research on this topic, we present variations in observational study findings and their validity, considering differences in population characteristics and data sources. We provide recommendations to enhance data quality for future research, including improvements in data recording platforms and strengthened structures for linking primary and secondary care electronic health records.

Chengyuan Hu, Zixin Lv, Jieping Zhu et al.

BackgroundLifestyle interventions are a critical component of weight loss programs, yet digital, personalized, and theory- and evidence-based lifestyle interventions remain limited. ObjectiveThis study aimed to investigate the effects of a combination of various dietary approaches and digital technology–based exercise on the body composition of college students with obesity. MethodsA total of 129 college students with obesity (mean age 18.3, SD 0.7 years; mean weight 89.9, SD 13.6 kg; mean BMI 30.6, SD 3.3 kg/m2) were initially recruited for this study. After excluding 2 participants, 127 students with obesity were ultimately included in the statistical analysis. An 8-week weight loss intervention was conducted with the students, combining exercise and various digitally supported dietary approaches. Body composition indicators (muscle mass and fat mass) were assessed before and after the intervention. Participants were divided into 3 experimental groups (twice-weekly fasting [TWF], low-calorie diet [LCD], and time-restricted feeding [TRF]). Between-group comparisons were made using a 1-way ANOVA, while within-group comparisons used a repeated-measures ANOVA. Linear mixed-effects models were used to examine the interaction effects between sex and time, as well as between sex and group. ResultsAll groups showed significant decreases in weight and BMI, and the TRF group also showed a significant decrease in BMI (P=.002), but there were significant sex differences. The male TWF group showed the largest decrease in weight (mean difference [MD] −4.86 kg; P<.001), BMI (MD −1.1 kg/m2; P<.001), visceral fat mass (MD −0.607 kg; P=.003) and subcutaneous fat mass (MD −1.987 kg; P<.001) at 8 weeks. Improvements in weight (MD −5.662 kg; P<.001) and BMI (MD −1.587 kg/m2; P<.001) were more pronounced in the LCD group of female participants (P<.001). Muscle mass declined significantly in male participants in the TRF group at 4 weeks (P<.001) but stabilized at 8 weeks (P=.87). Linear mixed effects models showed that the sex and diet interaction significantly affected subcutaneous fat mass (P=.02). The effect of TRF on muscle mass in male participants peaked at 4 weeks (P<.001), with no significant difference from the control group at 8 weeks (P=.91). ConclusionsThis study demonstrated that 3 diet-combined exercise regimens produced sex-specific improvements in body composition in college students with obesity. Male participants achieved maximum visceral fat mass loss after 8 weeks with TWF combined with exercise, whereas female participants achieved greater total body fat loss with LCD combined with exercise. The effectiveness of the closed-loop monitoring-feedback behavior modification was verified by digital technology through intelligent monitoring to improve dietary compliance and a real-time feedback mechanism to enhance the effect of the intervention. The sex and diet interaction significantly affected subcutaneous fat mass; women who used LCD and TRF needed additional protein supplementation. Digital technology shows great potential in obesity management and is worth promoting.

Sara Bosticardo, Matteo Battocchio, Mario Ocampo-Pineda et al.

Investigating myelin integrity within multiple sclerosis (MS) lesions and in normal-appearing white matter is crucial for understanding demyelination and remyelination processes. While most approaches assess global myelin changes or compare lesions with homologous regions in healthy controls, they do not allow direct within-tract comparisons between lesional and non-lesional tissue.We introduce the tractography-informed myelin estimate (TIME), a novel map designed to quantify tract-specific myelin loss. TIME integrates tractography with myelin-sensitive imaging, such as myelin volume fraction, to compare lesional and non-lesional segments within the same white matter tract. By modeling local deviations from the expected myelin volume fraction signal along streamlines, TIME captures tract-specific myelin damage while accounting for within-tract variability. TIME is based on a microstructure-informed tractography framework, with an extra compartment to model signal loss caused by lesions.We evaluated TIME in 159 MS patients, assessing its association with neurological disability at baseline and longitudinally over a median follow-up of two years. At baseline, higher myelin loss captured by TIME was significantly associated with worse disability (β = 0.14, p = 0.015). Longitudinally, greater baseline disability predicted faster TIME-quantified myelin loss, which was in turn associated with a higher risk of disability worsening. In contrast, lesion-averaged myelin volume fraction showed no significant associations with either baseline disability or its progression.TIME provides a detailed, tract-specific assessment of myelin damage, providing greater sensitivity than conventional metrics, highlighting its potential as a biomarker in MS.

Cosima Prahm, Michael Bressler, Tanja Gohlke et al.

Abstract Traumatic hand injuries frequently result in prolonged functional impairment. Adherence to conventional rehabilitation is frequently limited by pain, fatigue, and low engagement. Virtual reality systems may increase training volume, but many rely on handheld controllers and lack integration. StableHandVR is an immersive, gamified VR application using optical hand tracking for task-oriented functional hand and finger rehabilitation. In a single-center randomized controlled trial, 150 inpatients undergoing rehabilitation after traumatic hand injuries were assigned to either the StableHandVR intervention (n = 75) or an active control (n = 75) performing untargeted hand exercises during passive 360° VR exposure. Both groups completed 12 supervised sessions over three weeks. The primary outcome was active range of hand motion (ROM); secondary outcomes included thumb opposition (Kapandji), grip strength, upper-limb function (DASH), pain (NRS), quality of life (SF-36), usability (SUS), intrinsic motivation (IMI), and training adherence. The intervention group achieved significantly greater gains in wrist ROM (+27.8° vs. +17.3°; p < 0.001), and thumb opposition (p = 0.04). Pain during movement decreased in both groups. Patients using StableHandVR voluntarily exceeded the prescribed training volume by 63%, reporting higher perceived effort (p < 0.001), usefulness (p = 0.018), and excellent usability (SUS = 85.3). StableHandVR was found to enhance motor recovery, engagement, and adherence, supporting its integration into clinical rehabilitation pathways.

Amanda M. Lopez, Yoonjung Choi, Zhi Zhou

Chlorella is a promising biofuel source due to its high lipid accumulation, rapid growth, and suitability for inland cultivation. However, how the Paramecium bursaria Chlorella virus 1 (PBCV-1) influences its triacylglycerol (TAG) accumulation remains underexplored. This data article provides a detailed description of the dataset generated to investigate TAG accumulation profiles in Chlorella infected with PBCV-1. The data, collected via high-resolution epifluorescence microscopy of over 4000 single cells across a full lytic cycle, includes measurements of TAG accumulation, chlorophyll fluorescence, and nuclear morphology, along with extracellular nutrient concentrations to rule out nutrient stress as a confounding factor. This dataset can be reused by researchers to develop new image analysis algorithms, train machine learning models, investigate virus-host interactions, and inform the development of more cost-effective biofuel production strategies.

Ahmed Kabil, Ghada Khoriba, Mina Yousef et al.

Medical Image Segmentation (MIS) stands as a cornerstone in medical image analysis, playing a pivotal role in precise diagnostics, treatment planning, and monitoring of various medical conditions. This paper presents a comprehensive and systematic survey of MIS methodologies, bridging the gap between traditional image processing techniques and modern deep learning approaches. The survey encompasses thresholding, edge detection, region-based segmentation, clustering algorithms, and model-based techniques while also delving into state-of-the-art deep learning architectures such as Convolutional Neural Networks (CNNs), Fully Convolutional Networks (FCNs), and the widely adopted U-Net and its variants. Moreover, integrating attention mechanisms, semi-supervised learning, generative adversarial networks (GANs), and Transformer-based models is thoroughly explored. In addition to covering established methods, this survey highlights emerging trends, including hybrid architectures, cross-modality learning, federated and distributed learning frameworks, and active learning strategies, which aim to address challenges such as limited labeled datasets, computational complexity, and model generalizability across diverse imaging modalities. Furthermore, a specialized case study on lumbar spine segmentation is presented, offering insights into the challenges and advancements in this relatively underexplored anatomical region. Despite significant progress in the field, critical challenges persist, including dataset bias, domain adaptation, interpretability of deep learning models, and integration into real-world clinical workflows.

YunHe Su, Zhengyang Lu, Junhui Liu et al.

This paper explores the advancements and applications of large-scale models in the medical field, with a particular focus on Medical Large Models (MedLMs). These models, encompassing Large Language Models (LLMs), Vision Models, 3D Large Models, and Multimodal Models, are revolutionizing healthcare by enhancing disease prediction, diagnostic assistance, personalized treatment planning, and drug discovery. The integration of graph neural networks in medical knowledge graphs and drug discovery highlights the potential of Large Graph Models (LGMs) in understanding complex biomedical relationships. The study also emphasizes the transformative role of Vision-Language Models (VLMs) and 3D Large Models in medical image analysis, anatomical modeling, and prosthetic design. Despite the challenges, these technologies are setting new benchmarks in medical innovation, improving diagnostic accuracy, and paving the way for personalized healthcare solutions. This paper aims to provide a comprehensive overview of the current state and future directions of large models in medicine, underscoring their significance in advancing global health.

Faraz Farhadi, Shadi A. Esfahani, Fereshteh Yousefirizi et al.

The field of Clinical-Computational Nuclear Medicine is rapidly advancing, fueled by AI, tracer kinetic modeling, radiomics, and integrated informatics. These technologies improve imaging quality, automate lesion detection, and enable personalized radiopharmaceutical therapy through physiologically based pharmacokinetic (PBPK) modeling and voxel-level dosimetry. Workflow automation and Natural Language Processing (NLP) further enhance operational efficiency. However, successful implementation and adoption of these tools require clinical oversight to ensure accuracy, interpretability, and patient safety. This paper highlights key computational innovations and emphasizes the critical role of clinician-guided evaluation in shaping the future of precision imaging and therapy.

Kerol Djoumessi, Philipp Berens

Convolutional neural networks (CNNs) are widely used for high-stakes applications like medicine, often surpassing human performance. However, most explanation methods rely on post-hoc attribution, approximating the decision-making process of already trained black-box models. These methods are often sensitive, unreliable, and fail to reflect true model reasoning, limiting their trustworthiness in critical applications. In this work, we introduce SoftCAM, a straightforward yet effective approach that makes standard CNN architectures inherently interpretable. By removing the global average pooling layer and replacing the fully connected classification layer with a convolution-based class evidence layer, SoftCAM preserves spatial information and produces explicit class activation maps that form the basis of the model's predictions. Evaluated on three medical datasets, SoftCAM maintains classification performance while significantly improving both the qualitative and quantitative explanation compared to existing post-hoc methods. Our results demonstrate that CNNs can be inherently interpretable without compromising performance, advancing the development of self-explainable deep learning for high-stakes decision-making. The code is available at https://github.com/kdjoumessi/SoftCAM

Ryan B. Appleby, P. Basran

Artificial intelligence (AI) is a branch of computer science in which computer systems are designed to perform tasks that mimic human intelligence. Today, AI is reshaping day-to-day life and has numerous emerging medical applications poised to profoundly reshape the practice of veterinary medicine. In this Currents in One Health, we discuss the essential elements of AI for veterinary practitioners with the aim to help them make informed decisions in applying AI technologies into their practices. Veterinarians will play an integral role in ensuring the appropriate uses and good curation of data. The expertise of veterinary professionals will be vital to ensuring good data and, subsequently, AI that meets the needs of the profession. Readers interested in an in-depth description of AI and veterinary medicine are invited to explore a complementary manuscript of this Currents in One Health available in the May 2022 issue of the American Journal of Veterinary Research.

Bekzhan Faizov, Alizhan Bukayev, Zhaxylyk Sabitov et al.

This study presents a comprehensive analysis of 23 Y-STR data for the Merkit clan, a subgroup within the Kerey tribe of the Kazakh people. A total of 64 complete haplotypes were generated using the PowerPlex Y23 System. The data obtained using 23 Y-STR markers has been submitted to the Y Chromosome Haplotype Reference Database (YHRD) at yhrd.org, which will significantly enhance the forensic database for the Kazakh population in Kazakhstan. The research focuses on the distribution of haplotypes within the clan and their genealogical lines, which were visualized using a Median-joining network and Multidimensional scaling plot. The study identifies four distinct haplogroup clusters, revealing important insights into the genetic makeup and historical lineage of the Merkits. This dataset not only enriches our understanding of Kazakh genetic structure but also holds significant value for anthropological and population genetic research, as well as for forensic genetics. This work bridges a notable gap in genetic research on the Merkit clan, contributing to a deeper understanding of Central Asian nomadic tribes.

Davide Griffa, Alessio Natale, Yuri Merli et al.

<b>Introduction:</b> Chronic ulcers significantly burden healthcare systems, requiring precise measurement and assessment for effective treatment. Traditional methods, such as manual segmentation, are time-consuming and error-prone. This review evaluates the potential of artificial intelligence AI-powered mobile apps for automated ulcer segmentation and their application in clinical settings. <b>Methods:</b> A comprehensive literature search was conducted across PubMed, CINAHL, Cochrane, and Google Scholar databases. The review focused on mobile apps that use fully automatic AI algorithms for wound segmentation. Apps requiring additional hardware or needing more technical documentation were excluded. Vital technological features, clinical validation, and usability were analysed. <b>Results:</b> Ten mobile apps were identified, showing varying levels of segmentation accuracy and clinical validation. However, many apps did not publish sufficient information on the segmentation methods or algorithms used, and most lacked details on the databases employed for training their AI models. Additionally, several apps were unavailable in public repositories, limiting their accessibility and independent evaluation. These factors challenge their integration into clinical practice despite promising preliminary results. <b>Discussion:</b> AI-powered mobile apps offer significant potential for improving wound care by enhancing diagnostic accuracy and reducing the burden on healthcare professionals. Nonetheless, the lack of transparency regarding segmentation techniques, unpublished databases, and the limited availability of many apps in public repositories remain substantial barriers to widespread clinical adoption. <b>Conclusions:</b> AI-driven mobile apps for ulcer segmentation could revolutionise chronic wound management. However, overcoming limitations related to transparency, data availability, and accessibility is essential for their successful integration into healthcare systems.

Tianwei Zhang, Dong Wei, Mengmeng Zhu et al.

Self-supervised learning has emerged as a powerful tool for pretraining deep networks on unlabeled data, prior to transfer learning of target tasks with limited annotation. The relevance between the pretraining pretext and target tasks is crucial to the success of transfer learning. Various pretext tasks have been proposed to utilize properties of medical image data (e.g., three dimensionality), which are more relevant to medical image analysis than generic ones for natural images. However, previous work rarely paid attention to data with anatomy-oriented imaging planes, e.g., standard cardiac magnetic resonance imaging views. As these imaging planes are defined according to the anatomy of the imaged organ, pretext tasks effectively exploiting this information can pretrain the networks to gain knowledge on the organ of interest. In this work, we propose two complementary pretext tasks for this group of medical image data based on the spatial relationship of the imaging planes. The first is to learn the relative orientation between the imaging planes and implemented as regressing their intersecting lines. The second exploits parallel imaging planes to regress their relative slice locations within a stack. Both pretext tasks are conceptually straightforward and easy to implement, and can be combined in multitask learning for better representation learning. Thorough experiments on two anatomical structures (heart and knee) and representative target tasks (semantic segmentation and classification) demonstrate that the proposed pretext tasks are effective in pretraining deep networks for remarkably boosted performance on the target tasks, and superior to other recent approaches.

Yinchi Zhou, Tianqi Chen, Jun Hou et al.

Image-to-image translation is a vital component in medical imaging processing, with many uses in a wide range of imaging modalities and clinical scenarios. Previous methods include Generative Adversarial Networks (GANs) and Diffusion Models (DMs), which offer realism but suffer from instability and lack uncertainty estimation. Even though both GAN and DM methods have individually exhibited their capability in medical image translation tasks, the potential of combining a GAN and DM to further improve translation performance and to enable uncertainty estimation remains largely unexplored. In this work, we address these challenges by proposing a Cascade Multi-path Shortcut Diffusion Model (CMDM) for high-quality medical image translation and uncertainty estimation. To reduce the required number of iterations and ensure robust performance, our method first obtains a conditional GAN-generated prior image that will be used for the efficient reverse translation with a DM in the subsequent step. Additionally, a multi-path shortcut diffusion strategy is employed to refine translation results and estimate uncertainty. A cascaded pipeline further enhances translation quality, incorporating residual averaging between cascades. We collected three different medical image datasets with two sub-tasks for each dataset to test the generalizability of our approach. Our experimental results found that CMDM can produce high-quality translations comparable to state-of-the-art methods while providing reasonable uncertainty estimations that correlate well with the translation error.

Luming Chen, Yifan Qi, Aiping Wu et al.

Background. Chinese medical entities have not been organized comprehensively due to the lack of well-developed terminology systems, which poses a challenge to processing Chinese medical texts for fine-grained medical knowledge representation. To unify Chinese medical terminologies, mapping Chinese medical entities to their English counterparts in the Unified Medical Language System (UMLS) is an efficient solution. However, their mappings have not been investigated sufficiently in former research. In this study, we explore strategies for mapping Chinese medical entities to the UMLS and systematically evaluate the mapping performance. Methods. First, Chinese medical entities are translated to English using multiple web-based translation engines. Then, 3 mapping strategies are investigated: (a) string-based, (b) semantic-based, and (c) string and semantic similarity combined. In addition, cross-lingual pretrained language models are applied to map Chinese medical entities to UMLS concepts without translation. All of these strategies are evaluated on the ICD10-CN, Chinese Human Phenotype Ontology (CHPO), and RealWorld datasets. Results. The linear combination method based on the SapBERT and term frequency-inverse document frequency bag-of-words models perform the best on all evaluation datasets, with 91.85%, 82.44%, and 78.43% of the top 5 accuracies on the ICD10-CN, CHPO, and RealWorld datasets, respectively. Conclusions. In our study, we explore strategies for mapping Chinese medical entities to the UMLS and identify a satisfactory linear combination method. Our investigation will facilitate Chinese medical entity normalization and inspire research that focuses on Chinese medical ontology development.

Abel Centella-Artola, Arnoldo Bezanilla-Morlot, Roberto Serrano-Notivoli et al.

The paper presents a high-resolution (-3km) gridded dataset for daily precipitation across Cuba for 1961-2008, called CubaPrec1. The dataset was built using the information from the data series of 630 stations from the network operated by the National Institute of Water Resources. The original station data series were quality controlled using a spatial coherence process of the data, and the missing values were estimated on each day and location independently. Using the filled data series, a grid of 3 × 3 km spatial resolution was constructed by estimating daily precipitation and their corresponding uncertainties at each grid box. This new product represents a precise spatiotemporal distribution of precipitation in Cuba and provides a useful baseline for future studies in hydrology, climatology, and meteorology. The data collection described is available on zenodo: https://doi.org/10.5281/zenodo.7847844

Hongjian Zhou, Fenglin Liu, Boyang Gu et al.

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