Pamela M. Chiroque-Solano, M Lee Van Horn, Thomas Jaki
Precision medicine seeks to match patients with treatments that produce the greatest benefit. The Predicted Individual Treatment Effect (PITE)-the difference between predicted outcomes under treatment and control-quantifies this benefit but is difficult to estimate due to unobserved counterfactuals, high dimensionality, and complex interactions. We compared 30+ modeling strategies, including penalized and projection-based methods, flexible learners, and tree-ensembles, using a structured simulation framework varying sample size, dimensionality, multicollinearity, and interaction complexity. Performance was measured using root mean squared error (RMSE) for prediction accuracy and directional accuracy (DIR) for correctly classifying benefit versus harm. Internal validation produced optimistic estimates, whereas external validation with distributional shifts and higher-order interactions more clearly revealed model weaknesses. Penalized and projection-based approaches-ridge, lasso, elastic net, partial least squares (PLS), and principal components regression (PCR)-consistently achieved strong RMSE and DIR performance. Flexible learners excelled only under strong signals and sufficient sample sizes. Results highlight robust linear/projection defaults and the necessity of rigorous external validation.
Hematology analyzers are essential diagnostic and monitoring tools for detecting blood diseases. Although contemporary analyzers produce only basic insights, they are often not as detailed as required under the personalized medicine paradigm. Next-Generation Hematology Analyzers (NGHAs) are revolutionary newcomers in the field, with significant advantages over regular hematology analyzers. They provide deeper insights into cellular morphology, function, and genetic profiles. This detailed information opens up possibilities for tailor-made diagnostic and therapeutic approaches in precision medicine. This review presents some revolutionary technologies that have changed hematology analyzers and provides an overview of their limitations, basic functions, and influence on clinical practice. It focuses on the integration of state-of-the-art technologies, such as microfluidics, advanced optics, artificial intelligence, flow cytometry, and digital imaging, empowering NGHAs to improve diagnostic accuracy, rapidly detect diseases, and support flexible, targeted therapy. Hints regarding point-of-care hematology testing are also provided to discuss its implications for transforming healthcare patterns. This review highlights the data management, standardization, regulatory, and ethical challenges associated with these technologies. A review tracking the current state-of-the-art and trends for the future is provided to show how these advancements may reconfigure hematology analyzer design and act as a stepping stone for future therapeutic reforms.
Despite the success of large language models (LLMs) in various domains, their potential in Traditional Chinese Medicine (TCM) remains largely underexplored due to two critical barriers: (1) the scarcity of high-quality TCM data and (2) the inherently multimodal nature of TCM diagnostics, which involve looking, listening, smelling, and pulse-taking. These sensory-rich modalities are beyond the scope of conventional LLMs. To address these challenges, we present ShizhenGPT, the first multimodal LLM tailored for TCM. To overcome data scarcity, we curate the largest TCM dataset to date, comprising 100GB+ of text and 200GB+ of multimodal data, including 1.2M images, 200 hours of audio, and physiological signals. ShizhenGPT is pretrained and instruction-tuned to achieve deep TCM knowledge and multimodal reasoning. For evaluation, we collect recent national TCM qualification exams and build a visual benchmark for Medicinal Recognition and Visual Diagnosis. Experiments demonstrate that ShizhenGPT outperforms comparable-scale LLMs and competes with larger proprietary models. Moreover, it leads in TCM visual understanding among existing multimodal LLMs and demonstrates unified perception across modalities like sound, pulse, smell, and vision, paving the way toward holistic multimodal perception and diagnosis in TCM. Datasets, models, and code are publicly available. We hope this work will inspire further exploration in this field.
Mahmoud Alwakeel, Aditya Nagori, An-Kwok Ian Wong
et al.
Large Language Models have been tested on medical student-level questions, but their performance in specialized fields like Critical Care Medicine (CCM) is less explored. This study evaluated Meta-Llama 3.1 models (8B and 70B parameters) on 871 CCM questions. Llama3.1:70B outperformed 8B by 30%, with 60% average accuracy. Performance varied across domains, highest in Research (68.4%) and lowest in Renal (47.9%), highlighting the need for broader future work to improve models across various subspecialty domains.
Precision Medicine (PM) transforms the traditional "one-drug-fits-all" paradigm by customising treatments based on individual characteristics, and is an emerging topic for HCI research on digital health. A key element of PM, the Polygenic Risk Score (PRS), uses genetic data to predict an individual's disease risk. Despite its potential, PRS faces barriers to adoption, such as data inclusivity, psychological impact, and public trust. We conducted a mixed-methods study to explore how people perceive PRS, formed of surveys (n=254) and interviews (n=11) with UK-based participants. The interviews were supplemented by interactive storyboards with the ContraVision technique to provoke deeper reflection and discussion. We identified ten key barriers and five themes to PRS adoption and proposed design implications for a responsible PRS framework. To address the complexities of PRS and enhance broader PM practices, we introduce the term Human-Precision Medicine Interaction (HPMI), which integrates, adapts, and extends HCI approaches to better meet these challenges.
Recent advances in artificial intelligence (AI) and multimodal data collection are revolutionizing dermatology. Generative AI and machine learning approaches offer opportunities to enhance the diagnosis and treatment of inflammatory skin diseases, including atopic dermatitis, psoriasis, hidradenitis suppurativa, and autoimmune connective tissue disease. This review examines the current landscape of AI applications for inflammatory skin diseases and explores how generative AI and machine learning methods can advance the field through deep phenotyping, disease heterogeneity characterization, drug development, personalized medicine, and clinical care. We discuss the promises and challenges of these technologies and present a vision for their integration into clinical practice.
Gabriel Rotsen Fortes Aguiar, Geraldo Bezerra da Silva Júnior, Janaína de Almeida Mota Ramalho
et al.
Abstract Arboviruses are endemic in several countries and represent a worrying public health problem. The most important of these diseases is dengue fever, whose numbers continue to rise and have reached millions of annual cases in Brazil since the last decade. Other arboviruses of public health concern are chikungunya and Zika, both of which have caused recent epidemics, and yellow fever, which has also caused epidemic outbreaks in our country. Like most infectious diseases, arboviruses have the potential to affect the kidneys through several mechanisms. These include the direct action of the viruses, systemic inflammation, hemorrhagic phenomena and other complications, in addition to the toxicity of the drugs used in treatment. In this review article, the epidemiological aspects of the main arboviruses in Brazil and other countries where these diseases are endemic, clinical aspects and the main laboratory changes found, including changes in renal function, are addressed. It also describes how arboviruses behave in kidney transplant patients. The pathophysiological mechanisms of kidney injury associated with arboviruses are described and finally the recommended treatment for each disease and recommendations for kidney support in this context are given.
Objective The α-globin fusion gene between the HBA2 and HBAP1 genes, is clinically important in thalassemia screening because this fusion gene can cause severe hemoglobin (Hb) H disease when combined with α0 -thalassemia (α0 -thal). In this study, we evaluate the red blood cell parameters of α-thalassemia fusion gene in southern China.Method Study samples suspected of α-thalassemia fusion gene were collected and confirmed by PCR-sequencing from one medical lab center in southern China. Their genotypes and phenotypes were analyzed.Results A total of 266 cases of α-thalassemia fusion gene were confirmed in our lab from 2017 to 2023, most of them were from Hainan province (169 cases) and Huadu district of Guangzhou (21 cases), the nationality of 143 cases from Hainan was identified, with 71.3% (102/143) being from the Li minority. The Hb, MCV, MCH for αα/(αα)fusion in adult males were 143.5±11.83g/L, 81.51±4.39 fl, and 26.26±1.29 pg, respectively; and in females, they were 126.69±12.89 g/L, 80.10±4.05 fl, 25.8±2.04 pg, respectively. All 12 cases (αα) Fusion/ --SEA showed anemia with decreased Hb, MCV and MCH.Conclusion The carriers of α-globin fusion gene heterozygotes are clinically silent and exhibit an α+ phenotype. Individuals with (αα)Fusion/--SEA show apparent anemia. This α-globin fusion gene is relatively common in southern China, specifically among the Li minority of Hainan province. Therefore, it should be taken into account for genetic counseling purposes.
Andrew D. Leavitt, Johnny Mahlangu, Priyanka Raheja
et al.
Background: Valoctocogene roxaparvovec, an adeno-associated virus-mediated gene therapy for severe hemophilia A, enables endogenous factor (F)VIII expression and provides bleed protection. Objectives: Determine valoctocogene roxaparvovec durability, efficacy, and safety 4 years after treatment. Methods: In the phase 3 GENEr8-1 trial, 134 adult male persons with severe hemophilia A without inhibitors and previously using FVIII prophylaxis received a 6 × 1013 vg/kg infusion of valoctocogene roxaparvovec. Efficacy endpoints included annualized bleed rate, annualized FVIII infusion rate, FVIII activity, and the Haemophilia-Specific Quality of Life Questionnaire for Adults. Adverse events and immunosuppressant use were assessed. Change from baseline was assessed after participants discontinued prophylaxis (scheduled for week 4). Results: Median follow-up was 214.3 weeks; 2 participants discontinued since the previous data cutoff. Declines from baseline in mean treated annualized bleed rate (−82.6%; P < .0001) and annualized FVIII infusion rate (−95.5%; P < .0001) were maintained from previous years in the primary analysis population of 112 participants who enrolled from a noninterventional study. During year 4, 81 of 110 rollover participants experienced 0 treated bleeds. Week 208 mean and median chromogenic FVIII activity were 16.1 IU/dL and 6.7 IU/dL, respectively, in 130 modified intention-to-treat participants. Seven participants resumed prophylaxis since the previous data cutoff. Mean change from baseline to week 208 in Haemophilia-Specific Quality of Life Questionnaire for Adults Total Score (P < .0001) remained clinically meaningful for modified intention-to-treat participants. Alanine aminotransferase elevation was the most common adverse event during year 4 (56/131 participants); none required immunosuppressants. Conclusion: Valoctocogene roxaparvovec provides persistent FVIII expression, hemostatic control, and health-related quality of life improvements with no new safety signals.
Diabetes is an epidemic caused by a multitude of factors. Despite the studies attempting to unravel its mechanism, there is still more to discover about glucose–insulin dynamics. In a recent issue of the Journal of Clinical Investigation, Cheruiyot et al. uncovered a translational regulatory circuit during β‐cell glucose toxicity that inherently affects the translational makeup and protein expression in functioning β‐cells.Journal of Clinical Investigation, Cheruiyot et al. uncovered a translational regulatory circuit during β‐cell glucose toxicity that inherently affects the translational makeup and protein expression in functioning β‐cells. Their multiomics approach might provide a deeper understanding of high glucose and translational regulation of genes involved in β‐cell insulin impairment caused by prolonged high‐glucose exposure.
Diseases of the endocrine glands. Clinical endocrinology
Recently, with increasing interest in pet healthcare, the demand for computer-aided diagnosis (CAD) systems in veterinary medicine has increased. The development of veterinary CAD has stagnated due to a lack of sufficient radiology data. To overcome the challenge, we propose a generative active learning framework based on a variational autoencoder. This approach aims to alleviate the scarcity of reliable data for CAD systems in veterinary medicine. This study utilizes datasets comprising cardiomegaly radiograph data. After removing annotations and standardizing images, we employed a framework for data augmentation, which consists of a data generation phase and a query phase for filtering the generated data. The experimental results revealed that as the data generated through this framework was added to the training data of the generative model, the frechet inception distance consistently decreased from 84.14 to 50.75 on the radiograph. Subsequently, when the generated data were incorporated into the training of the classification model, the false positive of the confusion matrix also improved from 0.16 to 0.66 on the radiograph. The proposed framework has the potential to address the challenges of data scarcity in medical CAD, contributing to its advancement.
Manlio De Domenico, Luca Allegri, Guido Caldarelli
et al.
The adoption of digital twins (DTs) in precision medicine is increasingly viable, propelled by extensive data collection and advancements in artificial intelligence (AI), alongside traditional biomedical methodologies. However, the reliance on black-box predictive models, which utilize large datasets, presents limitations that could impede the broader application of DTs in clinical settings. We argue that hypothesis-driven generative models, particularly multiscale modeling, are essential for boosting the clinical accuracy and relevance of DTs, thereby making a significant impact on healthcare innovation. This paper explores the transformative potential of DTs in healthcare, emphasizing their capability to simulate complex, interdependent biological processes across multiple scales. By integrating generative models with extensive datasets, we propose a scenario-based modeling approach that enables the exploration of diverse therapeutic strategies, thus supporting dynamic clinical decision-making. This method not only leverages advancements in data science and big data for improving disease treatment and prevention but also incorporates insights from complex systems and network science, quantitative biology, and digital medicine, promising substantial advancements in patient care.
Afsar U. Ahmed, Saleh Almasabi, Ron Firestein
et al.
Colorectal cancer (CRC) is one of the most common forms of cancer worldwide and treatment options for advanced CRC, which has a low 5-year survival rate, remain limited. Integrin-linked kinase (ILK), a multifunctional, scaffolding, pseudo-kinase regulating many integrin-mediated cellular processes, is highly expressed in many cancers. However, the role of ILK in cancer progression is yet to be fully understood. We have previously uncovered a pro-inflammatory role for myeloid-specific ILK in dextran sodium sulfate (DSS)-induced colitis. To establish a correlation between chronic intestinal inflammation and colorectal cancer (CRC), we investigated the role of myeloid-ILK in mouse models of CRC. When myeloid-ILK deficient mice along with the WT control mice were subjected to colitis-associated and APCmin/+-driven CRC, tumour burden was reduced by myeloid-ILK deficiency in both models. The tumour-promoting phenotype of macrophages, M2 polarization, in vitro was impaired by the ILK deficiency and the number of M2-specific marker CD206-expressing tumour-associated macrophages (TAMs) in vivo were significantly diminished in myeloid-ILK deficient mice. Myeloid-ILK deficient mice showed enhanced tumour infiltration of CD8+ T cells and reduced tumour infiltration of FOXP3+ T cells in colitis-associated and APCmin/+-driven CRC, respectively, with an overall elevated CD8+/FOXP3+ ratio suggesting an anti-tumour immune phenotypes. In patient CRC tissue microarrays we observed elevated ILK+ myeloid (ILK+ CD11b+) cells in tumour sections compared to adjacent normal tissues, suggesting a conserved role for myeloid-ILK in CRC development in both human and animal models. This study identifies myeloid-specific ILK expression as novel driver of CRC, which could be targeted as a potential therapeutic option for advanced disease.
Persian Medicine (PM) uses wrist temperature/humidity and pulse to determine a person's health status and temperament. However, the diagnosis may depend on the physician's interpretation, hindering the combination of PM with modern medical methods. This study proposes a system for measuring pulse signals and temperament detection based on PM. The system uses recorded thermal distribution, a temperament questionnaire, and a customized pulse measurement device. The collected data can be sent to a physician via a telecare system for interpretation and prescription of medications. The system was clinically implemented for patient care, assessed the temperaments of 34 participants, and recorded thermal images of the wrist, back of the hand, and entire face. The study suggests that a customized device for measuring pulse waves and other criteria based on PM can be incorporated into a telemedicine system, reducing the dependency on PM specialists for diagnosis.
ObjectiveExtranodal natural killer/T cell lymphoma (NKTCL) is an aggressive EBV-related lymphoma, originating from NK cells or T cells. Previous study demonstrated that CD56 negative NKTCL should be recognized as a distinct subtype. In this study, the value of CD56 in NKTCL is validated in the era of asparaginase, and genomic analysis was done to dissect the differences between CD56-negative and positive NKTCL.Methods443 patients with newly diagnosed NKTCL were enrolled in this retrospective study, and correlation between CD56 positivity and survival outcomes was analyzed. The gene sequencing data was downloaded (http://www.biosino.org/node/project/detail/OEP000498), and bioinformatics analysis was done to delineate the tumor microenvironment and differentially expressed genes.ResultsCD56 was expressed in 337 patients (76.1%). Within a median follow-up time of 51 months, the 5-year overall survival (OS) and progression free survival (PFS) rates were 63.8% and 51.9%, respectively. For the whole cohort, patients who were CD56-positive had superior OS (5-year OS, 86.2% vs. 51.9%, p=0.019) and PFS (5-year PFS, 55.9% vs. 40.1%, p=0.016). For patients in early stage disease, CD56 positivity was associated with superior OS and PFS (p=0.008 and 0.005, respectively). In patients who received non-asparaginase-based chemotherapy, CD56-negative was associated with shorter OS and PFS (p<0.001), and in patients who received asparaginase-based chemotherapy, CD56-negative was not related to inferior OS and PFS (p=0.093 and p=0.829, respectively). The genomic analysis demonstrated that CD56 positive NKTCL probably originated from NK cells and CD56 negative NKTCL originated from T cells. CD56 positive NKTCL had significantly higher proportion of resting NK cells, activated NK cells, and activated CD8+ and CD4+ T cells in the tumor microenvironment.ConclusionsCD56 negative NKTCL differs from CD56 positive NKTCL in both the tumor microenvironment and survival outcomes, and asparaginase-based treatment may overcome the poor prognosis brought by CD56 negativity.