Development of machine-learning models using pharmacy inquiry database for predicting dose-related inquiries in a tertiary teaching hospital.

BACKGROUND: Drug-related problems (DRPs) are a significant concern in healthcare. Pharmacists play a vital role in detecting and resolving DRPs to improve patient safety. A pharmacy inquiry program was established in a tertiary teaching hospital to document inquiries about physicians' orders, aimed at preventing potential DRPs or providing medication information during order reviews. OBJECTIVE: We aimed to develop machine-learning models using a pharmacy inquiry database to predict dose-related inquiries based on prescriptions and patient information. METHODS: This retrospective study analyzed 20,393 pharmacy inquiries collected between January 2018 and February 2023. Data included prescription information (drug ingredient, dose, unit, and frequency), patient characteristics (age, sex, weight, and department), and renal function. The inquiries were categorized into two classes: dose-related inquiries (e.g., wrong dose and inappropriate regimen) and non-dose-related inquiries (e.g., inappropriate drug form and administration route). Six machine-learning models were developed: logistic regression, support vector classifier, decision tree, random forest, extreme gradient boosting, and categorical boosting. To evaluate the performance of the models, the area under the receiver operating characteristic curve and the accuracy were compared. RESULTS: The CatBoost model achieved the highest performance (sensitivity: 0.92; accuracy: 0.79). The SHapley Additive exPlanations values highlighted the importance of features in the model predictions, drug ingredients, units, and renal function, in that order. Notably, lower renal function positively contributed to the prediction of dose-related inquiries. Additionally, the subsequent feature importance among drug ingredients showed that drugs such as acetylsalicylic acid, famotidine, metformin, and spironolactone strongly influenced the prediction of dose-related inquiries. CONCLUSION: Machine-learning models that use pharmacy inquiry data can effectively predict dose-related inquiries. Further external validation and refinement of the models are required for broader applications in healthcare settings. These findings provide valuable guidance for healthcare professionals and highlight the potential of machine learning in pharmacists' decision-making.

Identifying facilitators of and barriers to the adoption of dynamic consent in digital health ecosystems: a scoping review.

BACKGROUND: Conventional consent practices face ethical challenges in continuously evolving digital health environments due to their static, one-time nature. Dynamic consent offers a promising solution, providing adaptability and flexibility to address these ethical concerns. However, due to the immaturity of the concept and accompanying technology, dynamic consent has not yet been widely used in practice. This study aims to identify the facilitators of and barriers to adopting dynamic consent in real-world scenarios. METHODS: This scoping review, conducted in December 2022, adhered to the PRISMA Extension for Scoping Reviews guidelines, focusing on dynamic consent within the health domain. A comprehensive search across Web of Science, PubMed, and Scopus yielded 22 selected articles based on predefined inclusion and exclusion criteria. RESULTS: The facilitators for the adoption of dynamic consent in digital health ecosystems were the provision of multiple consent modalities, personalized alternatives, continuous communication, and the dissemination of up-to-date information. Nevertheless, several barriers, such as consent fatigue, the digital divide, complexities in system implementation, and privacy and security concerns, needed to be addressed. This study also investigated current technological advancements and suggested considerations for further research aimed at resolving the remaining challenges surrounding dynamic consent. CONCLUSIONS: Dynamic consent emerges as an ethically advantageous method for digital health ecosystems, driven by its adaptability and support for continuous, two-way communication between data subjects and consumers. Ethical implementation in real-world settings requires the development of a robust technical framework capable of accommodating the diverse needs of stakeholders, thereby ensuring ethical integrity and data privacy in the evolving digital health landscape.

Deep learning-based diagnosis of lung cancer using a nationwide respiratory cytology image set: improving accuracy and inter-observer variability.

Deep learning (DL)-based image analysis has recently seen widespread application in digital pathology. Recent studies utilizing DL in cytopathology have shown promising results, however, the development of DL models for respiratory specimens is limited. In this study, we designed a DL model to improve lung cancer diagnosis accuracy using cytological images from the respiratory tract. This retrospective, multicenter study used digital cytology images of respiratory specimens from a quality-controlled national dataset collected from over 200 institutions. The image processing involves generating extended z-stack images to reduce the phase difference of cell clusters, color normalizing, and cropping image patches to 256 × 256 pixels. The accuracy of diagnosing lung cancer in humans from image patches before and after receiving AI assistance was compared. 30,590 image patches (1,273 whole slide images [WSIs]) were divided into 27,362 (1,146 WSIs) for training, 2,928 (126 WSIs) for validation, and 1,272 (1,272 WSIs) for testing. The Densenet121 model, which showed the best performance among six convolutional neural network models, was used for analysis. The results of sensitivity, specificity, and accuracy were 95.9%, 98.2%, and 96.9% respectively, outperforming the average of three experienced pathologists. The accuracy of pathologists after receiving AI assistance improved from 82.9% to 95.9%, and the inter-rater agreement of Fleiss' Kappa value was improved from 0.553 to 0.908. In conclusion, this study demonstrated that a DL model was effective in diagnosing lung cancer in respiratory cytology. By increasing diagnostic accuracy and reducing inter-observer variability, AI has the potential to enhance the diagnostic capabilities of pathologists.

3D microengineered vascularized tumor spheroids for drug delivery and efficacy testing.

Tumor angiogenesis is regarded as a promising target for limiting cancer progression because tumor-associated vasculature supplies blood and provides a path for metastasis. Thus, in vitro recapitulation of vascularized tumors is critical to understand the pathology of cancer and identify the mechanisms by which tumor cells proliferate, metastasize, and respond to drugs. In this study, we microengineered a vascularized tumor spheroid (VTS) model to reproduce the pathological features of solid tumors. We first generated tumor-EC hybrid spheroids with self-assembled intratumoral vessels, which enhanced the uniformity of the spheroids and peritumoral angiogenic capacity compared to spheroids composed only with cancer cells. Notably, the hybrid spheroids also exhibited expression profiles associated with aggressive behavior. The blood vessels sprouting around the hybrid spheroids on the VTS chip displayed the distinctive characteristics of leaky tumor vessels. With the VTS chip showing a progressive tumor phenotype, we validated the suppressive effects of axitinib on tumor growth and angiogenesis, which depended on exposure dose and time, highlighting the significance of tumor vascularization to predict the efficacy of anticancer drugs. Ultimately, we effectively induced both lymphangiogenesis and angiogenesis around the tumor spheroid by promoting interstitial flow. Thus, our VTS model is a valuable platform with which to investigate the interactions between tumor microenvironments and explore therapeutic strategies in cancer. STATEMENT OF SIGNIFICANCE: We conducted an integrative study within a vascularized tumor spheroid (VTS) model. We first generated tumor-EC hybrid spheroids with self-assembled intratumoral vessels, which enhanced the uniformity of the spheroids and peritumoral angiogenic capacity compared to spheroids composed only with cancer cells. Through RNA sequencing, we elucidated that the tumor-EC hybrid spheroids exhibited expression profiles associated with aggressive behavior such as cancer progression, invasion and metastasis. The blood vessels sprouting around the hybrid spheroids on the VTS chip displayed the distinctive characteristics of leaky tumor vessels. We further validated the suppressive effects of axitinib on tumor growth and angiogenesis, depending on exposure dose and time. Ultimately, we effectively induced both lymphangiogenesis and angiogenesis around the tumor spheroid by promoting interstitial flow.

Three-dimensional imaging and analysis of pathological tissue samples with de novo generation of citrate-based fluorophores.

Two-dimensional (2D) histopathology based on the observation of thin tissue slides is the current paradigm in diagnosis and prognosis. However, labeling strategies in conventional histopathology are limited in compatibility with 3D imaging combined with tissue clearing techniques. Here, we present a rapid and efficient volumetric imaging technique of pathological tissues called 3D tissue imaging through de novo formation of fluorophores, or 3DNFC, which is the integration of citrate-based fluorogenic reaction DNFC and tissue clearing techniques. 3DNFC markedly increases the fluorescence intensity of tissues by generating fluorophores on nonfluorescent amino-terminal cysteine and visualizes the 3D structure of the tissues to provide their anatomical morphology and volumetric information. Furthermore, the application of 3DNFC to pathological tissue achieves the 3D reconstruction for the unbiased analysis of diverse features of the disorders in their natural context. We suggest that 3DNFC is a promising volumetric imaging method for the prognosis and diagnosis of pathological tissues.

Development of a Frailty Detection Model Using Machine Learning with the Korean Frailty and Aging Cohort Study Data.

OBJECTIVES: This paper aimed to use machine learning to identify a new group of factors predicting frailty in the elderly population by utilizing the existing frailty criteria as a basis, as well as to validate the obtained results. METHODS: This study was conducted using data from the Korean Frailty and Aging Cohort Study (KFACS). The KFACS participants were classified as robust or frail based on Fried's frailty phenotype and excluded if they did not properly answer the questions, resulting in 1,066 robust and 165 frail participants. We then selected influential features through feature selection and trained the model using support vector machine, random forest, and gradient boosting algorithms with the prepared dataset. Due to the imbalanced distribution in the dataset with a low sample size, holdout was applied with stratified 10-fold and cross-validation for estimating the model performance. The reliability of the constructed model was validated using an unseen test set. The model was then trained with hyperparameter optimization. RESULTS: During the feature selection process, 27 features were identified as meaningful factors for frailty. The model was trained based on the selected features, and the weighted average F1-score reached 95.30% with the random forest algorithm. CONCLUSIONS: The results of the study demonstrated the possibility of adopting machine learning to strengthen existing frailty criteria. As the method analyzes questionnaire responses in a short time, it can support higher volumes of data on participants' health conditions and alert them regarding potential risks in advance.

A forebrain neural substrate for behavioral thermoregulation.

Thermoregulatory behavior is a basic motivated behavior for body temperature homeostasis. Despite its fundamental importance, a forebrain region or defined neural population required for this process has yet to be established. Here, we show that Vgat-expressing neurons in the lateral hypothalamus (LHVgat neurons) are required for diverse thermoregulatory behaviors. The population activity of LHVgat neurons is increased during thermoregulatory behavior and bidirectionally encodes thermal punishment and reward (P&R). Although this population also regulates feeding and caloric reward, inhibition of parabrachial inputs selectively impaired thermoregulatory behaviors and encoding of thermal stimulus by LHVgat neurons. Furthermore, two-photon calcium imaging revealed a subpopulation of LHVgat neurons bidirectionally encoding thermal P&R, which is engaged during thermoregulatory behavior, but is largely distinct from caloric reward-encoding LHVgat neurons. Our data establish LHVgat neurons as a required neural substrate for behavioral thermoregulation and point to the key role of the thermal P&R-encoding LHVgat subpopulation in thermoregulatory behavior.

Microfluidic device with brain extracellular matrix promotes structural and functional maturation of human brain organoids.

Brain organoids derived from human pluripotent stem cells provide a highly valuable in vitro model to recapitulate human brain development and neurological diseases. However, the current systems for brain organoid culture require further improvement for the reliable production of high-quality organoids. Here, we demonstrate two engineering elements to improve human brain organoid culture, (1) a human brain extracellular matrix to provide brain-specific cues and (2) a microfluidic device with periodic flow to improve the survival and reduce the variability of organoids. A three-dimensional culture modified with brain extracellular matrix significantly enhanced neurogenesis in developing brain organoids from human induced pluripotent stem cells. Cortical layer development, volumetric augmentation, and electrophysiological function of human brain organoids were further improved in a reproducible manner by dynamic culture in microfluidic chamber devices. Our engineering concept of reconstituting brain-mimetic microenvironments facilitates the development of a reliable culture platform for brain organoids, enabling effective modeling and drug development for human brain diseases.

3D Microfluidic Platform and Tumor Vascular Mapping for Evaluating Anti-Angiogenic RNAi-Based Nanomedicine.

Three-dimensional (3D) visualization of tumor vasculature is a key factor in accurate evaluation of RNA interference (RNAi)-based antiangiogenic nanomedicine, a promising approach for cancer therapeutics. However, this remains challenging because there is not a physiologically relevant in vitro model or precise analytic methodology. To address this limitation, a strategy based on 3D microfluidic angiogenesis-on-a-chip and 3D tumor vascular mapping was developed for evaluating RNAi-based antiangiogenic nanomedicine. We developed a microfluidic model to recapitulate functional 3D angiogenic sprouting when co-cultured with various cancer cell types. This model enabled efficient and rapid assessment of antiangiogenic nanomedicine in treatment of hyper-angiogenic cancer. In addition, tissue-clearing-based whole vascular mapping of tumor xenograft allowed extraction of complex 3D morphological information in diverse quantitative parameters. Using this 3D imaging-based analysis, we observed tumor sub-regional differences in the antiangiogenic effect. Our systematic strategy can help in narrowing down the promising targets of antiangiogenic nanomedicine and then enables deep analysis of complex morphological changes in tumor vasculature, providing a powerful platform for the development of safe and effective nanomedicine for cancer therapeutics.

Large-Scale 3D Optical Mapping and Quantitative Analysis of Nanoparticle Distribution in Tumor Vascular Microenvironment.

Nanoparticles (NPs) are a promising carrier for cancer therapeutics. Systemically administered NPs are transported to tumor tissues via the bloodstream, extravasated from microvessels, and delivered to cancer cells. The distribution of NPs in the tumor vascular microenvironment critically determines the therapeutic efficacy of NP-delivered drugs, but its precise assessment in 3D across a large volume remains challenging. Here, an analytical platform-termed OMNIA (for Optical Mapping of Nanoparticles and Image Analysis)-integrating tissue clearing, high-resolution optical imaging, and semiautomated image analysis is presented, which enables accurate, unbiased, and quantitative analysis of the distribution of NPs in relation to the vasculature across a large 3D volume. Application of OMNIA to tumor tissues revealed higher accumulation and more efficient extravasation of NPs in the tumor periphery than the core. Time-course analysis demonstrated that the accumulation of NPs in tumor peaked at 24 h after injection, but the relative distribution of NPs from the vasculature remained remarkably stable over time. Comparisons between 45- and 200-nm-sized NPs showed a lower accumulation of smaller NPs in tumors relative to the liver, yet better vessel permeation. Together, our results demonstrate that OMNIA facilitates precise and reliable evaluation of NP biodistribution, and mechanistic investigations on NP delivery to tumor tissues.