Medical big data: promise and challenges

Comparing neural language models for medical concept representation and patient trajectory prediction

Effective representation of medical concepts is crucial for secondary analyses of electronic health records. Neural language models have shown promise in automatically deriving medical concept representations from clinical data. However, the comparative performance of different language models for creating these empirical representations, and the extent to which they encode medical semantics, has not been extensively studied. This study aims to address this gap by evaluating the effectiveness of three popular language models - word2vec, fastText, and GloVe - in creating medical concept embeddings that capture their semantic meaning. By using a large dataset of digital health records, we created patient trajectories and used them to train the language models. We then assessed the ability of the learned embeddings to encode semantics through an explicit comparison with biomedical terminologies, and implicitly by predicting patient outcomes and trajectories with different levels of available information. Our qualitative analysis shows that empirical clusters of embeddings learned by fastText exhibit the highest similarity with theoretical clustering patterns obtained from biomedical terminologies, with a similarity score between empirical and theoretical clusters of 0.88, 0.80, and 0.92 for diagnosis, procedure, and medication codes, respectively. Conversely, for outcome prediction, word2vec and GloVe tend to outperform fastText, with the former achieving AUROC as high as 0.78, 0.62, and 0.85 for length-of-stay, readmission, and mortality prediction, respectively. In predicting medical codes in patient trajectories, GloVe achieves the highest performance for diagnosis and medication codes (AUPRC of 0.45 and of 0.81, respectively) at the highest level of the semantic hierarchy, while fastText outperforms the other models for procedure codes (AUPRC of 0.66). Our study demonstrates that subword information is crucial for learning medical concept representations, but global embedding vectors are better suited for more high-level downstream tasks, such as trajectory prediction. Thus, these models can be harnessed to learn representations that convey clinical meaning, and our insights highlight the potential of using machine learning techniques to semantically encode medical data.

Development and validation of an explainable model of brain injury in premature infants: A prospective cohort study

BACKGROUND

Preterm brain injury (PBI) is a prevalent complication in preterm infants, leading to the destruction of critical structural and functional brain connections and placing a significant burden on families. The timely detection of PBI is of paramount importance for the prevention and treatment of the condition. However, the absence of specific clinical manifestations in the early stages of PBI renders it susceptible to misdiagnosis and missed diagnoses. Moreover, once it occurs, there is no specific treatment available. The aim of this study was to develop and validate a machine learning (ML) based interpretable model for the early detection of PBI, as well as the assessment of patient-wide and individual risk factors for this disease.

METHODS

This study utilized a cohort of premature infants provided by Northwest Women's and Children's Hospital in China, comprising medical records of 650 premature infants, spanning from 2019 to 2021. PBI were identified based on cranial magnetic resonance imaging (MRI). Fourteen machine learning models were employed with stratified 10-fold cross-validation method used to evaluate model performance. The Shapley Additive Explanations (SHAP) method was applied for model interpretation. Feature selection methods were used to determine the final model which was validated on the independent test set. Subsequently, risk factors for the entire cohort and individual patients were assessed.

RESULTS

Among the fourteen machine learning models, the CatBoost model demonstrated the best discriminative ability. Following feature selection, the final model was constructed using seven features, designated as PBIPred (Preterm Brain Injury Predictor). PBIPred exhibited strong performance in both 10-fold cross-validation and independent test set (AUC = 0.8229) for accurately predicting PBI. The screening for risk factors in the cohort and individuals identified the following variables as positive risk factors for PBI: Mechanical ventilation (MV), Weight, Anemia of prematurity (AOP), Respiratory distress syndrome (RDS), Albumin (ALB), and White blood cell (WBC).

AVAILABILITY AND IMPLEMENTATION

The PBIPred webserver and PBIPred tool were developed for clinical diagnosis and large-scale local medical record data prediction. They can be accessed freely at http://pbipred.liaolab.net and https://github.com/chikit2077/PBIPred, respectively.

Clinical validation and optimization of machine learning models for early prediction of sepsis

Introduction

Sepsis is a global health threat that has a high incidence and mortality rate. Early prediction of sepsis onset can drive effective interventions and improve patients' outcome.

Methods

Data were collected retrospectively from a cohort of 2,329 adult patients with positive bacteria cultures from a tertiary hospital in China between October 1, 2019 and September 30, 2020. Thirty six clinical features were selected as inputs for the models. We trained models in predicting sepsis by machine learning (ML) methods, including logistic regression, decision tree, random forest (RF), multi-layer perceptron, and light gradient boosting. We evaluated the performance of the five ML models and the evaluation metrics were: area under the ROC curve (AUC), accuracy, F1-score, sensitivity and specificity. The data of another cohort of 2,286 patients between October 1, 2020 and April 1, 2022 were used to validate the performance of the model performing best in the in the internal validation set. Shapley additive explanations (SHAP) method was applied to evaluate feature importance and explain the predictions of this model.

Results

Of the five machine learning models developed, the RF model demonstrated the best performance in terms of AUC (0.818), F1 value (0.38), and sensitivity (0.746). The RF model also has a comparable AUC (0.771) in the external validation set. The SHAP method identified procalcitonin, albumin, prothrombin time, and sex as the important variables contributing to the prediction of sepsis.

Discussion

The RF model we developed showed the greatest potential for early prediction of sepsis in admitted patients, which could aid clinicians in their decision-making process. Our findings also suggested that male patients with bacterial infections and high procalcitonin levels, lower albumin levels, or prolonged prothrombin times were more likely to develop sepsis.

Drug repositioning for Parkinson's disease: An emphasis on artificial intelligence approaches

Parkinson's disease (PD) is one of the most incapacitating neurodegenerative diseases (NDDs). PD is the second most common NDD worldwide which affects approximately 1-2 percent of people over 65 years. It is an attractive pursuit for artificial intelligence (AI) to contribute to and evolve PD treatments through drug repositioning by repurposing existing drugs, shelved drugs, or even candidates that do not meet the criteria for clinical trials. A search was conducted in three databases Web of Science, Scopus, and PubMed. We reviewed the data related to the last years (1975-present) to identify those drugs currently being proposed for repositioning in PD. Moreover, we reviewed the present status of the computational approach, including AI/Machine Learning (AI/ML)-powered pharmaceutical discovery efforts and their implementation in PD treatment. It was found that the number of drug repositioning studies for PD has increased recently. Repositioning of drugs in PD is taking off, and scientific communities are increasingly interested in communicating its results and finding effective treatment alternatives for PD. A better chance of success in PD drug discovery has been made possible due to AI/ML algorithm advancements. In addition to the experimentation stage of drug discovery, it is also important to leverage AI in the planning stage of clinical trials to make them more effective. New AI-based models or solutions that increase the success rate of drug development are greatly needed.

Computer vision algorithms in healthcare: Recent advancements and future challenges

Computer vision has emerged as a promising technology with numerous applications in healthcare. This systematic review provides an overview of advancements and challenges associated with computer vision in healthcare. The review highlights the application areas where computer vision has made significant strides, including medical imaging, surgical assistance, remote patient monitoring, and telehealth. Additionally, it addresses the challenges related to data quality, privacy, model interpretability, and integration with existing healthcare systems. Ethical and legal considerations, such as patient consent and algorithmic bias, are also discussed. The review concludes by identifying future directions and opportunities for research, emphasizing the potential impact of computer vision on healthcare delivery and outcomes. Overall, this systematic review underscores the importance of understanding both the advancements and challenges in computer vision to facilitate its responsible implementation in healthcare.

Body Surface Potential Mapping: A Perspective on High-Density Cutaneous Electrophysiology

The electrophysiological signals recorded by cutaneous electrodes, known as body surface potentials (BSPs), are widely employed biomarkers in medical diagnosis. Despite their widespread application and success in detecting various conditions, the poor spatial resolution of traditional BSP measurements poses a limit to their diagnostic potential. Advancements in the field of bioelectronics have facilitated the creation of compact, high-quality, high-density recording arrays for cutaneous electrophysiology, allowing detailed spatial information acquisition as BSP maps (BSPMs). Currently, the design of electrode arrays for BSP mapping lacks a standardized framework, leading to customizations for each clinical study, limiting comparability, reproducibility, and transferability. This perspective proposes preliminary design guidelines, drawn from existing literature, rooted solely in the physical properties of electrophysiological signals and mathematical principles of signal processing. These guidelines aim to simplify and generalize the optimization process for electrode array design, fostering more effective and applicable clinical research. Moreover, the increased spatial information obtained from BSPMs introduces interpretation challenges. To mitigate this, two strategies are outlined: observational transformations that reconstruct signal sources for intuitive comprehension, and machine learning-driven diagnostics. BSP mapping offers significant advantages in cutaneous electrophysiology with respect to classic electrophysiological recordings and is expected to expand into broader clinical domains in the future.

The impacts on population health by China's regional health data centers and the potential mechanism of influence

Background

China recently established a series of pilot regional health data centers with a mandate to acquire, consolidate, analyze, and translate data into evidence for health policy decision-making. This experiment with "big data" has the potential to influence population health and is the focus of this study.

Methods

This study used national longitudinal survey data from the China Family Panel Studies over the period 2014-2020 to empirically assess the impact of China's establishment of pilot regional health data centers on population health and health inequality. A difference-in-differences model was employed to investigate the policy effect on population health, with additional exploration of possible mechanisms of influence. The corrected concentration index was used to measure health inequality, while Wagstaff decomposition method was applied to examine the marginal influence of the policy effect on health inequality.

Results

Overall health status of local residents has improved after the establishment of the pilot regional health data centers. Using mechanism analysis, the findings demonstrated that improvements to population health were driven by promoting healthy lifestyles and innovations in medical practices. Furthermore, due to differences in individual e-health literacy, the pilot centers produced "pro-rich" health inequality where high-income groups benefited more from the establishment of the pilot centers in terms of health than low-income groups.

Conclusions

This study has highlighted the potential to improve population health, in general, with the advent of big data centers, but for these benefits be unevenly distributed among the resident population.

3D-Printed Polymeric Biomaterials for Health Applications

3D printing, also known as additive manufacturing, holds immense potential for rapid prototyping and customized production of functional health-related devices. With advancements in polymer chemistry and biomedical engineering, polymeric biomaterials have become integral to 3D-printed biomedical applications. However, there still exists a bottleneck in the compatibility of polymeric biomaterials with different 3D printing methods, as well as intrinsic challenges such as limited printing resolution and rates. Therefore, this review aims to introduce the current state-of-the-art in 3D-printed functional polymeric health-related devices. It begins with an overview of the landscape of 3D printing techniques, followed by an examination of commonly used polymeric biomaterials. Subsequently, examples of 3D-printed biomedical devices are provided and classified into categories such as biosensors, bioactuators, soft robotics, energy storage systems, self-powered devices, and data science in bioplotting. The emphasis is on exploring the current capabilities of 3D printing in manufacturing polymeric biomaterials into desired geometries that facilitate device functionality and studying the reasons for material choice. Finally, an outlook with challenges and possible improvements in the near future is presented, projecting the contribution of general 3D printing and polymeric biomaterials in the field of healthcare.

Information Mode-Dependent Success Rates of Obtaining German Medical Informatics Initiative-Compliant Broad Consent in the Emergency Department: Single-Center Prospective Observational Study

BACKGROUND

The broad consent (BC) developed by the German Medical Informatics Initiative is a pivotal national strategy for obtaining patient consent to use routinely collected data from electronic health records, insurance companies, contact information, and biomaterials for research. Emergency departments (EDs) are ideal for enrolling diverse patient populations in research activities. Despite regulatory and ethical challenges, obtaining BC from patients in ED with varying demographic, socioeconomic, and disease characteristics presents a promising opportunity to expand the availability of ED data.

OBJECTIVE

This study aimed to evaluate the success rate of obtaining BC through different consenting approaches in a tertiary ED and to explore factors influencing consent and dropout rates.

METHODS

A single-center prospective observational study was conducted in a German tertiary ED from September to December 2022. Every 30th patient was screened for eligibility. Eligible patients were informed via one of three modalities: (1) directly in the ED, (2) during their inpatient stay on the ward, or (3) via telephone after discharge. The primary outcome was the success rate of obtaining BC within 30 days of ED presentation. Secondary outcomes included analyzing potential influences on the success and dropout rates based on patient characteristics, information mode, and the interaction time required for patients to make an informed decision.

RESULTS

Of 11,842 ED visits, 419 patients were screened for BC eligibility, with 151 meeting the inclusion criteria. Of these, 68 (45%) consented to at least 1 BC module, while 24 (15.9%) refused participation. The dropout rate was 39.1% (n=59) and was highest in the telephone-based group (57/109, 52.3%) and lowest in the ED group (1/14, 7.1%). Patients informed face-to-face during their inpatient stay following the ED treatment had the highest consent rate (23/27, 85.2%), while those approached in the ED or by telephone had consent rates of 69.2% (9/13 and 36/52). Logistic regression analysis indicated that longer interaction time significantly improved consent rates (P=.03), while female sex was associated with higher dropout rates (P=.02). Age, triage category, billing details (inpatient treatment), or diagnosis did not significantly influence the primary outcome (all P>.05).

CONCLUSIONS

Obtaining BC in an ED environment is feasible, enabling representative inclusion of ED populations. However, discharge from the ED and female sex negatively affected consent rates to the BC. Face-to-face interaction proved most effective, particularly for inpatients, while telephone-based approaches resulted in higher dropout rates despite comparable consent rates to direct consenting in the ED. The findings underscore the importance of tailored consent strategies and maintaining consenting staff in EDs and on the wards to enhance BC information delivery and consent processes for eligible patients.

TRIAL REGISTRATION

German Clinical Trials Register DRKS00028753; https://drks.de/search/de/trial/DRKS00028753.

Synthetic Breast Ultrasound Images: A Study to Overcome Medical Data Sharing Barriers

The vast potential of medical big data to enhance healthcare outcomes remains underutilized due to privacy concerns, which restrict cross-center data sharing and the construction of diverse, large-scale datasets. To address this challenge, we developed a deep generative model aimed at synthesizing medical data to overcome data sharing barriers, with a focus on breast ultrasound (US) image synthesis. Specifically, we introduce CoLDiT, a conditional latent diffusion model with a transformer backbone, to generate US images of breast lesions across various Breast Imaging Reporting and Data System (BI-RADS) categories. Using a training dataset of 9,705 US images from 5,243 patients across 202 hospitals with diverse US systems, CoLDiT generated breast US images without duplicating private information, as confirmed through nearest-neighbor analysis. Blinded reader studies further validated the realism of these images, with area under the receiver operating characteristic curve (AUC) scores ranging from 0.53 to 0.77. Additionally, synthetic breast US images effectively augmented the training dataset for BI-RADS classification, achieving performance comparable to that using an equal-sized training set comprising solely real images (P = 0.81 for AUC). Our findings suggest that synthetic data, such as CoLDiT-generated images, offer a viable, privacy-preserving solution to facilitate secure medical data sharing and advance the utilization of medical big data.

TMODINET: A trustworthy multi-omics dynamic learning integration network for cancer diagnostic

Multiple types of omics data contain a wealth of biomedical information which reflect different aspects of clinical samples. Multi-omics integrated analysis is more likely to lead to more accurate clinical decisions. Existing cancer diagnostic methods based on multi-omics data integration mainly focus on the classification accuracy of the model, while neglecting the interpretability of the internal mechanism and the reliability of the results, which are crucial in specific domains such as precision medicine and the life sciences. To overcome this limitation, we propose a trustworthy multi-omics dynamic learning framework (TMODINET) for cancer diagnostic. The framework employs multi-omics adaptive dynamic learning to process each sample to provide patient-centered personality diagnosis by using self-attentional learning of features and modalities. To characterize the correlation between samples well, we introduce a graph dynamic learning method which can adaptively adjust the graph structure according to the specific classification results for specific graph convolutional networks (GCN) learning. Moreover, we utilize an uncertainty mechanism by employing Dirichlet distribution and Dempster-Shafer theory to obtain uncertainty and integrate multi-omics data at the decision level, ensuring trustworthy for cancer diagnosis. Extensive experiments on four real-world multimodal medical datasets are conducted. Compared to state-of-the-art methods, the superior performance and trustworthiness of our proposed algorithm are clearly validated. Our model has great potential for clinical diagnosis.

HERALD: A domain-specific query language for longitudinal health data analytics

BACKGROUND

Large-scale health data has significant potential for research and innovation, especially with longitudinal data offering insights into prevention, disease progression, and treatment effects. Yet, analyzing this data type is complex, as data points are repeatedly documented along the timeline. As a consequence, extracting cross-sectional tabular data suitable for statistical analysis and machine learning can be challenging for medical researchers and data scientists alike, with existing tools lacking balance between ease-of-use and comprehensiveness.

OBJECTIVE

This paper introduces HERALD, a novel domain-specific query language designed to support the transformation of longitudinal health data into cross-sectional tables. We describe the basic concepts, the query syntax, a graphical user interface for constructing and executing HERALD queries, as well as an integration into Informatics for Integrating Biology and the Bedside (i2b2).

METHODS

The syntax of HERALD mimics natural language and supports different query types for selection, aggregation, analysis of relationships, and searching for data points based on filter expressions and temporal constraints. Using a hierarchical concept model, queries are executed individually for the data of each patient, while constructing tabular output. HERALD is closed, meaning that queries process data points and generate data points. Queries can refer to data points that have been produced by previous queries, providing a simple, but powerful nesting mechanism.

RESULTS

The open-source implementation consists of a HERALD query parser, an execution engine, as well as a web-based user interface for query construction and statistical analysis. The implementation can be deployed as a standalone component and integrated into self-service data analytics environments like i2b2 as a plugin. HERALD can be valuable tool for data scientists and machine learning experts, as it simplifies the process of transforming longitudinal health data into tables and data matrices.

CONCLUSION

The construction of cross-sectional tables from longitudinal data can be supported through dedicated query languages that strike a reasonable balance between language complexity and transformation capabilities.

Business intelligence systems for population health management: a scoping review

Objective

Population health management (PHM) is a promising data-driven approach to address the challenges faced by health care systems worldwide. Although Business Intelligence (BI) systems are known to be relevant for a data-driven approach, the usage for PHM is limited in its elaboration. To explore available scientific publications, a systematic review guided by PRISMA was conducted of mature BI initiatives to investigate their decision contexts and BI capabilities.

Materials and Methods

PubMed, Embase, and Web of Science were searched for articles published from January 2012 through November 2023. Articles were included if they described a (potential) BI system for PHM goals. Additional relevant publications were identified through snowballing. Technological Readiness Levels were evaluated to select mature initiatives from the 29 initiatives found. From the 11 most mature systems the decision context (eg, patient identification, risk stratification) and BI capabilities (eg, data warehouse, linked biobank) were extracted.

Results

The initiatives found are highly fragmented in decision context and BI capabilities. Varied terminology is used and much information is missing. Impact on population's health is currently limited for most initiatives. Care Link, CommunityRx, and Gesundes Kinzigtal currently stand out in aligning BI capabilities with their decision contexts.

Discussion and Conclusion

PHM is a data-driven approach that requires a coherent data strategy and understanding of decision contexts and user needs. Effective BI capabilities depend on this understanding. Designing public-private partnerships to protect intellectual property while enabling rapid knowledge development is crucial. Development of a framework is proposed for systematic knowledge building.

Detection of Disease Features on Retinal OCT Scans Using RETFound

Eye diseases such as age-related macular degeneration (AMD) are major causes of irreversible vision loss. Early and accurate detection of these diseases is essential for effective management. Optical coherence tomography (OCT) imaging provides clinicians with in vivo, cross-sectional views of the retina, enabling the identification of key pathological features. However, manual interpretation of OCT scans is labor-intensive and prone to variability, often leading to diagnostic inconsistencies. To address this, we leveraged the RETFound model, a foundation model pretrained on 1.6 million unlabeled retinal OCT images, to automate the classification of key disease signatures on OCT. We finetuned RETFound and compared its performance with the widely used ResNet-50 model, using single-task and multitask modes. The dataset included 1770 labeled B-scans with various disease features, including subretinal fluid (SRF), intraretinal fluid (IRF), drusen, and pigment epithelial detachment (PED). The performance was evaluated using accuracy and AUC-ROC values, which ranged across models from 0.75 to 0.77 and 0.75 to 0.80, respectively. RETFound models display comparable specificity and sensitivity to ResNet-50 models overall, making it also a promising tool for retinal disease diagnosis. These findings suggest that RETFound may offer improved diagnostic accuracy and interpretability for specific tasks, potentially aiding clinicians in more efficient and reliable OCT image analysis.

Digital therapeutics in Korea: current status, challenges, and future directions - a narrative review

Digital therapeutics (DTx) are emerging as a transformative innovation in healthcare offering evidence-based digital interventions for the treatment, management, and prevention of various diseases and disorders. In Korea, DTx have gained significant attention as potential solutions to the increasing burden of chronic diseases and mental health conditions. However, the Korean DTx market faces several challenges that hinder its widespread adoption and integration into the national healthcare system. This study provides a comprehensive analysis of the current state of the DTx market in Korea, identifies the key challenges impeding its growth, and proposes strategies for overcoming these obstacles. This study utilized a literature review and market analysis approach to examine the latest research, industry reports, and regulatory documents related to DTx. The analysis focused on three primary areas: (1) the current regulatory landscape, (2) technological advancements and challenges, and (3) economic and commercial factors influencing DTx adoption in Korea. A comparative analysis of global regulatory practices was also conducted to identify best practices. The findings revealed that while Korea has made significant strides in supporting DTx development, the market remains in its early stages. The key challenges include underdeveloped regulatory frameworks, issues with data quality and security, and a lack of established reimbursement pathways. We recommend developing tailored regulatory frameworks for DTx, enhancing policy support for small and medium-sized enterprises involved in DTx development, and increasing investments in technological infrastructure. By addressing these challenges, Korea could position itself as a leader in the global DTx market, delivering innovative and effective treatments to enhance patient care and outcomes.

Exploring barriers and ethical challenges to medical data sharing: perspectives from Chinese researchers

BACKGROUND

The impetus for policies promoting medical data sharing in China has gained significant traction. Nonetheless, the present legal and ethical framework governing the research use of medical data in China, is characterized by a more restrictive rather than permissive approach. The proportion of Chinese medical data being leveraged for scientific research still has room for improvement at present, indicating a significant untapped potential for advancing medical knowledge and improving healthcare outcomes. Building upon this research, we aim to delve deeper into the challenges researchers encounter in the sharing of medical data through focus group interviews.

METHODS

We conducted two focus group interviews study with researchers representing diverse disciplines to explore their perspectives on 21 June 2021 and 28 July 2021. A total of seventeen researchers willingly participated in this study, representing various professional backgrounds. Similar codes were merged. Research team discussions were also utilized to select interviewees' statements that were regarded as typical or representative.

RESULTS

The respondents demonstrated a strong understanding that medical data should not be disseminated arbitrarily, recognizing the importance of sharing data in compliance with laws. Through the interview, we found that although respondents stressed the importance of careful consideration regarding if and when this information can be responsibly released, none of the respondents raised the issue of necessitating consent from data subjects for the research use of medical data. This observation sharply contrasts with the stringent separate consent provisions for secondary data use outlined in the PIPL.

CONCLUSIONS

The findings from the focus group studies shed light on researchers' barriers and ethical challenges towards medical data sharing for scientific research, highlighting their deep concern for data security and cautious approach to sharing. The key objectives aimed at facilitating and enabling the reuse of medical data encompass enhancing interoperability, harmonizing data standards, improving data quality, safeguarding privacy, ensuring informed consent, incentivizing patients, and establishing explicit regulations pertaining to data access and utilization.

Identifying Risk of Postoperative Cardiorespiratory Complications in OSA

BACKGROUND

Patients with OSA are at increased risk of postoperative cardiorespiratory complications and death. Attempts to stratify this risk have been inadequate, and predictors from large, well-characterized cohort studies are needed.

RESEARCH QUESTION

What is the relationship between OSA severity, defined by various polysomnography-derived metrics, and risk of postoperative cardiorespiratory complications or death, and which metrics best identify such risk?

STUDY DESIGN AND METHODS

In this cohort study, 6,770 consecutive patients who underwent diagnostic polysomnography for possible OSA and a procedure involving general anesthesia within a period of 2 years before and at least 5 years after polysomnography. Participants were identified by linking polysomnography and health databases. Relationships between OSA severity measures and the composite primary outcome of cardiorespiratory complications or death within 30 days of hospital discharge were investigated using univariable and multivariable analyses.

RESULTS

The primary outcome was observed in 5.3% (n = 361) of the cohort. Although univariable analysis showed strong dose-response relationships between this outcome and multiple OSA severity measures, multivariable analysis showed its independent predictors were: age older than 65 years (OR, 2.67 [95% CI, 2.03-3.52]; P < .0001), age 55.1 to 65 years (OR, 1.47 [95% CI, 1.09-1.98]; P = .0111), time between polysomnography and procedure of ≥ 5 years (OR, 1.32 [95% CI, 1.02-1.70]; P = .0331), BMI of ≥ 35 kg/m2 (OR, 1.43 [95% CI, 1.13-1.82]; P = .0032), presence of known cardiorespiratory risk factor (OR, 1.63 [95% CI, 1.29-2.06]; P < .0001), > 4.7% of sleep time at an oxygen saturation measured by pulse oximetry of < 90% (T90; OR, 1.91 [95% CI, 1.51-2.42]; P < .0001), and cardiothoracic procedures (OR, 7.95 [95% CI, 5.71-11.08]; P < .0001). For noncardiothoracic procedures, age, BMI, presence of known cardiorespiratory risk factor, and percentage of sleep time at an oxygen saturation of < 90% remained the significant predictors, and a risk score based on their ORs was predictive of outcome (area under receiver operating characteristic curve, 0.7 [95% CI, 0.64-0.75]).

INTERPRETATION

These findings provide a basis for better identifying high-risk patients with OSA and determining appropriate postoperative care.

Integrated Approach Using Intuitionistic Fuzzy Multicriteria Decision-Making to Support Classifier Selection for Technology Adoption in Patients with Parkinson Disease: Algorithm Development and Validation

Background

Parkinson disease (PD) is reported to be among the most prevalent neurodegenerative diseases globally, presenting ongoing challenges and increasing burden on health care systems. In an effort to support patients with PD, their carers, and the wider health care sector to manage this incurable condition, the focus has begun to shift away from traditional treatments. One of the most contemporary treatments includes prescribing assistive technologies (ATs), which are viewed as a way to promote independent living and deliver remote care. However, the uptake of these ATs is varied, with some users not ready or willing to accept all forms of AT and others only willing to adopt low-technology solutions. Consequently, to manage both the demands on resources and the efficiency with which ATs are deployed, new approaches are needed to automatically assess or predict a user's likelihood to accept and adopt a particular AT before it is prescribed. Classification algorithms can be used to automatically consider the range of factors impacting AT adoption likelihood, thereby potentially supporting more effective AT allocation. From a computational perspective, different classification algorithms and selection criteria offer various opportunities and challenges to address this need.

Objective

This paper presents a novel hybrid multicriteria decision-making approach to support classifier selection in technology adoption processes involving patients with PD.

Methods

First, the intuitionistic fuzzy analytic hierarchy process (IF-AHP) was implemented to calculate the relative priorities of criteria and subcriteria considering experts' knowledge and uncertainty. Second, the intuitionistic fuzzy decision-making trial and evaluation laboratory (IF-DEMATEL) was applied to evaluate the cause-effect relationships among criteria/subcriteria. Finally, the combined compromise solution (CoCoSo) was used to rank the candidate classifiers based on their capability to model the technology adoption.

Results

We conducted a study involving a mobile smartphone solution to validate the proposed methodology. Structure (F5) was identified as the factor with the highest relative priority (overall weight=0.214), while adaptability (F4) (D-R=1.234) was found to be the most influencing aspect when selecting classifiers for technology adoption in patients with PD. In this case, the most appropriate algorithm for supporting technology adoption in patients with PD was the A3 - J48 decision tree (M3=2.5592). The results obtained by comparing the CoCoSo method in the proposed approach with 2 alternative methods (simple additive weighting and technique for order of preference by similarity to ideal solution) support the accuracy and applicability of the proposed methodology. It was observed that the final scores of the algorithms in each method were highly correlated (Pearson correlation coefficient >0.8).

Conclusions

The IF-AHP-IF-DEMATEL-CoCoSo approach helped to identify classification algorithms that do not just discriminate between good and bad adopters of assistive technologies within the Parkinson population but also consider technology-specific features like design, quality, and compatibility that make these classifiers easily implementable by clinicians in the health care system.

Big data analytics in the healthcare sector: Opportunities and challenges in developing countries. A literature review

Background: Despite the ongoing efforts to digitalize the healthcare sector in developing countries, the full adoption of big data analytics in healthcare settings is yet to be attained Exploring opportunities and challenges encountered is essential for designing and implementing effective interventional strategies. Objective: Exploring opportunities and challenges towards integrating big data analytics technologies in the healthcare industry in developing countries. Methodology: This was a narrative review study design. A literature search on different databases was conducted including PubMed, ScienceDirect, MEDLINE, Scopus, and Google Scholar. Articles with predetermined keywords and written in English were included. Results: Big data analytics finds its application in population health management and clinical decision-support systems even in developing countries. The major challenges towards the integration of big data analytics in the healthcare sector in developing countries include fragmentation of healthcare data and lack of interoperability, data security, privacy and confidentiality concerns, limited resources and inadequate regulatory and policy frameworks for governing big data analytics technologies and limited reliable power and internet infrastructures. Conclusion: Digitalization of healthcare delivery in developing countries faces several significant challenges. However, the integration of big data analytics can potentially open new avenues for enhancing healthcare delivery with cost-effective benefits.

Evolving Concepts of Pain Management in Elderly Patients

PURPOSE OF REVIEW

The elderly population typically suffer from a variety of diseases that mostly reflect the degenerative changes linked with the aging process. These diseases may be exacerbated by acute pain or by an abrupt aggravation of previously stable chronic pain.

RECENT FINDINGS

Physical and psychological changes associated with aging may influence one's experience of pain and, as a result, the severity of pain. Pain treatment in the elderly can be complex and is often a budgetary burden on the nation's health care system. These difficulties arise, in part, because of unanticipated pharmacodynamics, changed pharmacokinetics, and polypharmacy interactions. Therefore, it is critical to integrate a multidisciplinary team to develop a management strategy that incorporates medical, psychological, and surgical methods to control persistent pain conditions. It is in this critical process that pain prediction models can be of great use. The purpose of pain prediction models for the elderly is the use of mathematical models to predict the occurrence and intensity of pain and pain-related conditions. These mathematical models employ a vast quantity of data to ascertain the many risk factors for the development of pain problems in the elderly, whether said risks are adjustable or not. These models will pave the way for more informed medical decision making that are based on the findings of thousands of patients who have previously experienced the same illness and related pain conditions. However, future additional research needs to be undertaken to build prediction models that are not constrained by substantial legal or methodological limitations.

An Edge-Cloud-Aided Private High-Order Fuzzy C-Means Clustering Algorithm in Smart Healthcare

Smart healthcare has emerged to provide healthcare services using data analysis techniques. Especially, clustering is playing an indispensable role in analyzing healthcare records. However, large multi-modal healthcare data imposes great challenges on clustering. Specifically, it is hard for traditional approaches to obtain desirable results for healthcare data clustering since they are not able to work for multi-modal data. This paper presents a new high-order multi-modal learning approach using multimodal deep learning and the Tucker decomposition (F- HoFCM). Furthermore, we propose an edge-cloud-aided private scheme to facilitate the clustering efficiency for its embedding in edge resources. Specifically, the computationally intensive tasks, such as parameter updating with high-order back propagation algorithm and clustering through high-order fuzzy c-means, are processed in a centralized location with cloud computing. The other tasks such as multi-modal data fusion and Tucker decomposition are performed at the edge resources. Since the feature fusion and Tucker decomposition are nonlinear operations, the cloud cannot obtain the raw data, thus protecting the privacy. Experimental results state that the presented approach produces significantly more accurate results than the existing high-order fuzzy c-means (HOFCM) on multi-modal healthcare datasets and furthermore the clustering efficiency are significantly improved by the developed edge-cloud-aided private healthcare system.

Baseline Data and Measurement Instruments Reported in Observational Studies in Inflammatory Bowel Disease: Results from a Systematic Review

BACKGROUND

Heterogeneity in demographic and outcomes data with corresponding measurement instruments [MIs] creates barriers to data pooling and analysis. Several core outcome sets have been developed in inflammatory bowel disease [IBD] to homogenize outcomes data. A parallel Minimum Data Set [MDS] for baseline characteristics is lacking. We conducted a systematic review to develop the first MDS.

METHODS

A systematic review was made of observational studies from three databases [2000-2021]. Titles and abstracts were screened, full-text articles were reviewed, and data were extracted by two reviewers. Baseline data were grouped into ten domains: demographics, clinical features, disease behaviour/complications, biomarkers, endoscopy, histology, radiology, healthcare utilization and patient-reported data. Frequency of baseline data and MIs within respective domains are reported.

RESULTS

From 315 included studies [600 552 subjects], most originated from Europe [196; 62%] and North America [59; 19%], and were published between 2011 and 2021 [251; 80%]. The most frequent domains were demographics [311; 98.7%] and clinical [289; 91.7%]; 224 [71.1%] studies reported on the triad of sex [306; 97.1%], age [289; 91.7%], and disease phenotype [231; 73.3%]. Few included baseline data for radiology [19; 6%], healthcare utilization [19; 6%], and histology [17; 5.4%]. Ethnicity [19; 6%], race [17; 5.4%], and alcohol/drug consumption [6; 1.9%] were the least reported demographics. From 25 MIs for clinical disease activity, the Harvey-Bradshaw Index [n = 53] and Mayo score [n = 37] were most frequently used.

CONCLUSIONS

Substantial variability exists in baseline population data reporting. These findings will inform a future consensus for MDS in IBD to enhance data harmonization and credibility of real-world evidence.

Data-Driven Identification of Factors That Influence the Quality of Adverse Event Reports: 15-Year Interpretable Machine Learning and Time-Series Analyses of VigiBase and QUEST

BACKGROUND

The completeness of adverse event (AE) reports, crucial for assessing putative causal relationships, is measured using the vigiGrade completeness score in VigiBase, the World Health Organization global database of reported potential AEs. Malaysian reports have surpassed the global average score (approximately 0.44), achieving a 5-year average of 0.79 (SD 0.23) as of 2019 and approaching the benchmark for well-documented reports (0.80). However, the contributing factors to this relatively high report completeness score remain unexplored.

OBJECTIVE

This study aims to explore the main drivers influencing the completeness of Malaysian AE reports in VigiBase over a 15-year period using vigiGrade. A secondary objective was to understand the strategic measures taken by the Malaysian authorities leading to enhanced report completeness across different time frames.

METHODS

We analyzed 132,738 Malaysian reports (2005-2019) recorded in VigiBase up to February 2021 split into historical International Drug Information System (INTDIS; n=63,943, 48.17% in 2005-2016) and newer E2B (n=68,795, 51.83% in 2015-2019) format subsets. For machine learning analyses, we performed a 2-stage feature selection followed by a random forest classifier to identify the top features predicting well-documented reports. We subsequently applied tree Shapley additive explanations to examine the magnitude, prevalence, and direction of feature effects. In addition, we conducted time-series analyses to evaluate chronological trends and potential influences of key interventions on reporting quality.

RESULTS

Among the analyzed reports, 42.84% (56,877/132,738) were well documented, with an increase of 65.37% (53,929/82,497) since 2015. Over two-thirds (46,186/68,795, 67.14%) of the Malaysian E2B reports were well documented compared to INTDIS reports at 16.72% (10,691/63,943). For INTDIS reports, higher pharmacovigilance center staffing was the primary feature positively associated with being well documented. In recent E2B reports, the top positive features included reaction abated upon drug dechallenge, reaction onset or drug use duration of <1 week, dosing interval of <1 day, reports from public specialist hospitals, reports by pharmacists, and reaction duration between 1 and 6 days. In contrast, reports from product registration holders and other health care professionals and reactions involving product substitution issues negatively affected the quality of E2B reports. Multifaceted strategies and interventions comprising policy changes, continuity of education, and human resource development laid the groundwork for AE reporting in Malaysia, whereas advancements in technological infrastructure, pharmacovigilance databases, and reporting tools concurred with increases in both the quantity and quality of AE reports.

CONCLUSIONS

Through interpretable machine learning and time-series analyses, this study identified key features that positively or negatively influence the completeness of Malaysian AE reports and unveiled how Malaysia has developed its pharmacovigilance capacity via multifaceted strategies and interventions. These findings will guide future work in enhancing pharmacovigilance and public health.

Advances in the study of tertiary lymphoid structures in the immunotherapy of breast cancer

Breast cancer, as one of the most common malignancies in women, exhibits complex and heterogeneous pathological characteristics across different subtypes. Triple-negative breast cancer (TNBC) and HER2-positive breast cancer are two common and highly invasive subtypes within breast cancer. The stability of the breast microbiota is closely intertwined with the immune environment, and immunotherapy is a common approach for treating breast cancer.Tertiary lymphoid structures (TLSs), recently discovered immune cell aggregates surrounding breast cancer, resemble secondary lymphoid organs (SLOs) and are associated with the prognosis and survival of some breast cancer patients, offering new avenues for immunotherapy. Machine learning, as a form of artificial intelligence, has increasingly been used for detecting biomarkers and constructing tumor prognosis models. This article systematically reviews the latest research progress on TLSs in breast cancer and the application of machine learning in the detection of TLSs and the study of breast cancer prognosis. The insights provided contribute valuable perspectives for further exploring the biological differences among different subtypes of breast cancer and formulating personalized treatment strategies.

Machine learning models with time-series clinical features to predict radiographic progression in patients with ankylosing spondylitis

Objective

Ankylosing spondylitis (AS) is chronic inflammatory arthritis causing structural damage and radiographic progression to the spine due to repeated and continuous inflammation over a long period. This study establishes the application of machine learning models to predict radiographic progression in AS patients using time-series data from electronic medical records (EMRs).

Methods

EMR data, including baseline characteristics, laboratory findings, drug administration, and modified Stoke AS Spine Score (mSASSS), were collected from 1,123 AS patients between January 2001 and December 2018 at a single center at the time of first (T1), second (T2), and third (T3) visits. The radiographic progression of the (n+1)th visit (Pn+1=(mSASSSn+1-mSASSSn)/(Tn+1-Tn)≥1 unit per year) was predicted using follow-up visit datasets from T1 to Tn. We used three machine learning methods (logistic regression with the least absolute shrinkage and selection operation, random forest, and extreme gradient boosting algorithms) with three-fold cross-validation.

Results

The random forest model using the T1 EMR dataset best predicted the radiographic progression P2 among the machine learning models tested with a mean accuracy and area under the curves of 73.73% and 0.79, respectively. Among the T1 variables, the most important variables for predicting radiographic progression were in the order of total mSASSS, age, and alkaline phosphatase.

Conclusion

Prognosis predictive models using time-series data showed reasonable performance with clinical features of the first visit dataset when predicting radiographic progression.

HyperTMO: a trusted multi-omics integration framework based on hypergraph convolutional network for patient classification

MOTIVATION

The rapid development of high-throughput biomedical technologies can provide researchers with detailed multi-omics data. The multi-omics integrated analysis approach based on machine learning contributes a more comprehensive perspective to human disease research. However, there are still significant challenges in representing single-omics data and integrating multi-omics information.

RESULTS

This article presents HyperTMO, a Trusted Multi-Omics integration framework based on Hypergraph convolutional network for patient classification. HyperTMO constructs hypergraph structures to represent the association between samples in single-omics data, then evidence extraction is performed by hypergraph convolutional network, and multi-omics information is integrated at an evidence level. Last, we experimentally demonstrate that HyperTMO outperforms other state-of-the-art methods in breast cancer subtype classification and Alzheimer's disease classification tasks using multi-omics data from TCGA (BRCA) and ROSMAP datasets. Importantly, HyperTMO is the first attempt to integrate hypergraph structure, evidence theory, and multi-omics integration for patient classification. Its accurate and robust properties bring great potential for applications in clinical diagnosis.

AVAILABILITY AND IMPLEMENTATION

HyperTMO and datasets are publicly available at https://github.com/ippousyuga/HyperTMO.

Revealing the Complexity of Fatigue: A Review of the Persistent Challenges and Promises of Artificial Intelligence

Part I reviews persistent challenges obstructing progress in understanding complex fatigue's biology. Difficulties quantifying subjective symptoms, mapping multi-factorial mechanisms, accounting for individual variation, enabling invasive sensing, overcoming research/funding insularity, and more are discussed. Part II explores how emerging artificial intelligence and machine and deep learning techniques can help address limitations through pattern recognition of complex physiological signatures as more objective biomarkers, predictive modeling to capture individual differences, consolidation of disjointed findings via data mining, and simulation to explore interventions. Conversational agents like Claude and ChatGPT also have potential to accelerate human fatigue research, but they currently lack capacities for robust autonomous contributions. Envisioned is an innovation timeline where synergistic application of enhanced neuroimaging, biosensors, closed-loop systems, and other advances combined with AI analytics could catalyze transformative progress in elucidating fatigue neural circuitry and treating associated conditions over the coming decades.

Rethinking the pros and cons of randomized controlled trials and observational studies in the era of big data and advanced methods: a panel discussion

Randomized controlled trials (RCTs) have traditionally been considered the gold standard for medical evidence. However, in light of emerging methodologies in data science, many experts question the role of RCTs. Within this context, experts in the USA and Canada came together to debate whether the primacy of RCTs as the gold standard for medical evidence, still holds in light of recent methodological advances in data science and in the era of big data. The purpose of this manuscript, aims to raise awareness of the pros and cons of RCTs and observational studies in order to help guide clinicians, researchers, students, and decision-makers in making informed decisions on the quality of medical evidence to support their work. In particular, new and underappreciated advantages and disadvantages of both designs are contrasted. Innovations taking place in both of these research methodologies, which can blur the lines between the two, are also discussed. Finally, practical guidance for clinicians and future directions in assessing the quality of evidence is offered.

Toward the Internet of Medical Things: Architecture, trends and challenges

In recent years, the growing pervasiveness of wearable technology has created new opportunities for medical and emergency rescue operations to protect users' health and safety, such as cost-effective medical solutions, more convenient healthcare and quick hospital treatments, which make it easier for the Internet of Medical Things (IoMT) to evolve. The study first presents an overview of the IoMT before introducing the IoMT architecture. Later, it portrays an overview of the core technologies of the IoMT, including cloud computing, big data and artificial intelligence, and it elucidates their utilization within the healthcare system. Further, several emerging challenges, such as cost-effectiveness, security, privacy, accuracy and power consumption, are discussed, and potential solutions for these challenges are also suggested.

Design and technical validation to generate a synthetic 12-lead electrocardiogram dataset to promote artificial intelligence research

Purpose

The purpose of this study is to construct a synthetic dataset of ECG signal that overcomes the sensitivity of personal information and the complexity of disclosure policies.

Methods

The public dataset was constructed by generating synthetic data based on the deep learning model using a convolution neural network (CNN) and bi-directional long short-term memory (Bi-LSTM), and the effectiveness of the dataset was verified by developing classification models for ECG diagnoses.

Results

The synthetic 12-lead ECG dataset generated consists of a total of 6000 ECGs, with normal and 5 abnormal groups. The synthetic ECG signal has a waveform pattern similar to the original ECG signal, the average RMSE between the two signals is 0.042 µV, and the average cosine similarity is 0.993. In addition, five classification models were developed to verify the effect of the synthetic dataset and showed performance similar to that of the model made with the actual dataset. In particular, even when the real dataset was applied as a test set to the classification model trained with the synthetic dataset, the classification performance of all models showed high accuracy (average accuracy 93.41%).

Conclusion

The synthetic 12-lead ECG dataset was confirmed to perform similarly to the real-world 12-lead ECG in the classification model. This implies that a synthetic dataset can perform similarly to a real dataset in clinical research using AI. The synthetic dataset generation process in this study provides a way to overcome the medical data disclosure challenges constrained by privacy rights, a way to encourage open data policies, and contribute significantly to promoting cardiovascular disease research.

Understanding patient needs and predicting outcomes in IgA nephropathy using data analytics and artificial intelligence: a narrative review

This narrative review explores two case scenarios related to immunoglobulin A nephropathy (IgAN) and the application of predictive monitoring, big data analysis and artificial intelligence (AI) in improving treatment outcomes. The first scenario discusses how online service providers accurately understand consumer preferences and needs through the use of AI-powered big data analysis. The author, a clinical nephrologist, contemplates the potential application of similar methodologies, including AI, in his medical practice to better understand and meet patient needs. The second scenario presents a case study of a 20-year-old man with IgAN. The patient exhibited recurring symptoms, including gross haematuria and tonsillitis, over a 2-year period. Through histological examination and treatment with renin-angiotensin system blockade and corticosteroids, the patient experienced significant improvement in kidney function and reduced proteinuria over 15 years of follow-up. The case highlights the importance of individualized treatment strategies and the use of predictive tools, such as AI-based predictive models, in assessing treatment response and predicting long-term outcomes in IgAN patients. The article further discusses the collection and analysis of real-world big data, including electronic health records, for studying disease natural history, predicting treatment responses and identifying prognostic biomarkers. Challenges in integrating data from various sources and issues such as missing data and data processing limitations are also addressed. Mathematical models, including logistic regression and Cox regression analysis, are discussed for predicting clinical outcomes and analysing changes in variables over time. Additionally, the application of machine learning algorithms, including AI techniques, in analysing big data and predicting outcomes in IgAN is explored. In conclusion, the article highlights the potential benefits of leveraging AI-powered big data analysis, predictive monitoring and machine learning algorithms to enhance patient care and improve treatment outcomes in IgAN.

LeafAI: query generator for clinical cohort discovery rivaling a human programmer

OBJECTIVE

Identifying study-eligible patients within clinical databases is a critical step in clinical research. However, accurate query design typically requires extensive technical and biomedical expertise. We sought to create a system capable of generating data model-agnostic queries while also providing novel logical reasoning capabilities for complex clinical trial eligibility criteria.

MATERIALS AND METHODS

The task of query creation from eligibility criteria requires solving several text-processing problems, including named entity recognition and relation extraction, sequence-to-sequence transformation, normalization, and reasoning. We incorporated hybrid deep learning and rule-based modules for these, as well as a knowledge base of the Unified Medical Language System (UMLS) and linked ontologies. To enable data-model agnostic query creation, we introduce a novel method for tagging database schema elements using UMLS concepts. To evaluate our system, called LeafAI, we compared the capability of LeafAI to a human database programmer to identify patients who had been enrolled in 8 clinical trials conducted at our institution. We measured performance by the number of actual enrolled patients matched by generated queries.

RESULTS

LeafAI matched a mean 43% of enrolled patients with 27 225 eligible across 8 clinical trials, compared to 27% matched and 14 587 eligible in queries by a human database programmer. The human programmer spent 26 total hours crafting queries compared to several minutes by LeafAI.

CONCLUSIONS

Our work contributes a state-of-the-art data model-agnostic query generation system capable of conditional reasoning using a knowledge base. We demonstrate that LeafAI can rival an experienced human programmer in finding patients eligible for clinical trials.

Using Omaha System data to explore relationships between client outcomes, phenotypes, and targeted home intervention approaches: an exemplar examining practice effectiveness for older women with circulation problems

BACKGROUND

Improved health among older women remains elusive and may be linked to limited knowledge of and interventions targeted to population subgroups. Use of structured community nurse home visit data exploring relationships between client outcomes, phenotypes, and targeted intervention approaches may reveal new understandings of practice effectiveness.

MATERIALS AND METHODS

Omaha System data of 2363 women 65 years and older with circulation problems receiving at least 2 community nurse home visits were accessed. Previously identified phenotypes (Poor circulation; Irregular heart rate; and Limited symptoms), 7 intervention approaches (High-Surveillance; High-Teaching/Guidance/Counseling; Balanced-All; Balanced-Surveillance-Teaching/Guidance/Counseling; Low-Teaching/Guidance/Counseling-Balanced Other; Low-Surveillance-Mostly-Teaching/Guidance/Couseling-TreatmentProcedure-CaseManagement; and Mostly-TreatementProcedure+CaseManagement), and client knowledge, behavior, and status outcomes were used. Client-linked intervention approach counts, proportional use per phenotypes, and associations with client outcome scores were descriptively analyzed. Associations between intervention approach proportional use by phenotype and outcome scores were analyzed using parallel coordinate graph methodology for intervention approach effectiveness.

RESULTS

Percent use of intervention approach differed significantly by phenotype. The 2 most widely employed intervention approaches were characterized by either a high use of surveillance interventions or a balanced use of all intervention categories (surveillance, teaching/guidance/counseling, treatment-procedure, case-management). Mean outcome discharge and change scores significantly differed by intervention approach. Proportionally deployed intervention approach patterns by phenotype were associated with outcome small effects improvement.

DISCUSSIONS AND CONCLUSIONS

The Omaha System taxonomy supported the management and exploration of large multidimensional community nursing data of older women with circulation problems. This study offers a new way to examine intervention effectiveness using phenotype- and targeted intervention approach-informed structured data.

A call for better validation of opioid overdose risk algorithms

Clinical decision support (CDS) systems powered by predictive models have the potential to improve the accuracy and efficiency of clinical decision-making. However, without sufficient validation, these systems have the potential to mislead clinicians and harm patients. This is especially true for CDS systems used by opioid prescribers and dispensers, where a flawed prediction can directly harm patients. To prevent these harms, regulators and researchers have proposed guidance for validating predictive models and CDS systems. However, this guidance is not universally followed and is not required by law. We call on CDS developers, deployers, and users to hold these systems to higher standards of clinical and technical validation. We provide a case study on two CDS systems deployed on a national scale in the United States for predicting a patient's risk of adverse opioid-related events: the Stratification Tool for Opioid Risk Mitigation (STORM), used by the Veterans Health Administration, and NarxCare, a commercial system.

"In Situ" Joint Preparation Technique for First Metatarsophalangeal Arthrodesis: A Retrospective Comparative Review of 388 Cases

"Cup-shaped power reamers" and "flat cuts" (FC) are common joint preparation techniques in first metatarsophalangeal (MTP) joint arthrodesis. However, the third option of an "in situ" (IS) technique has rarely been studied. This study aims to compare the clinical, radiographic, and patient-reported outcomes (PROMs) of the IS technique for various MTP pathologies with other MTP joint preparation techniques. A single-center retrospective review was performed for patients who underwent primary MTP joint arthrodesis between 2015 and 2019. In total, 388 cases were included in the study. We found higher nonunion rates in the IS group (11.1% vs 4.6%, p = .016). However, the revision rates were similar between the groups (7.1% vs 6.5%, p = .809). Multivariate analysis revealed that diabetes mellitus was associated with significantly higher overall complication rates (p < .001). The FC technique was associated with transfer metatarsalgia (p = .015) and a more first ray shortening (p < .001). Visual analog scale, PROMIS-10 physical, and PROMIS-CAT physical scores significantly improved in IS and FC groups (p < .001, p = .002, p = .001, respectively). The improvement was comparable between the joint preparation techniques (p = .806). In conclusion, the IS joint preparation technique is simple and effective for first MTP joint arthrodesis. In our series, the IS technique had a higher radiographic nonunion rate that did not correlate with a higher revision rate, and otherwise similar complication profile to the FC technique while providing similar PROMs. The IS technique resulted in significantly less first ray shortening when compared to the FC technique.

Is There a Role of Artificial Intelligence in Preclinical Imaging?

This review provides an overview of the current opportunities for integrating artificial intelligence methods into the field of preclinical imaging research in nuclear medicine. The growing demand for imaging agents and therapeutics that are adapted to specific tumor phenotypes can be excellently served by the evolving multiple capabilities of molecular imaging and theranostics. However, the increasing demand for rapid development of novel, specific radioligands with minimal side effects that excel in diagnostic imaging and achieve significant therapeutic effects requires a challenging preclinical pipeline: from target identification through chemical, physical, and biological development to the conduct of clinical trials, coupled with dosimetry and various pre, interim, and post-treatment staging images to create a translational feedback loop for evaluating the efficacy of diagnostic or therapeutic ligands. In virtually all areas of this pipeline, the use of artificial intelligence and in particular deep-learning systems such as neural networks could not only address the above-mentioned challenges, but also provide insights that would not have been possible without their use. In the future, we expect that not only the clinical aspects of nuclear medicine will be supported by artificial intelligence, but that there will also be a general shift toward artificial intelligence-assisted in silico research that will address the increasingly complex nature of identifying targets for cancer patients and developing radioligands.

Handling missing values in healthcare data: A systematic review of deep learning-based imputation techniques

OBJECTIVE

The proper handling of missing values is critical to delivering reliable estimates and decisions, especially in high-stakes fields such as clinical research. In response to the increasing diversity and complexity of data, many researchers have developed deep learning (DL)-based imputation techniques. We conducted a systematic review to evaluate the use of these techniques, with a particular focus on the types of data, intending to assist healthcare researchers from various disciplines in dealing with missing data.

MATERIALS AND METHODS

We searched five databases (MEDLINE, Web of Science, Embase, CINAHL, and Scopus) for articles published prior to February 8, 2023 that described the use of DL-based models for imputation. We examined selected articles from four perspectives: data types, model backbones (i.e., main architectures), imputation strategies, and comparisons with non-DL-based methods. Based on data types, we created an evidence map to illustrate the adoption of DL models.

RESULTS

Out of 1822 articles, a total of 111 were included, of which tabular static data (29%, 32/111) and temporal data (40%, 44/111) were the most frequently investigated. Our findings revealed a discernible pattern in the choice of model backbones and data types, for example, the dominance of autoencoder and recurrent neural networks for tabular temporal data. The discrepancy in imputation strategy usage among data types was also observed. The "integrated" imputation strategy, which solves the imputation task simultaneously with downstream tasks, was most popular for tabular temporal data (52%, 23/44) and multi-modal data (56%, 5/9). Moreover, DL-based imputation methods yielded a higher level of imputation accuracy than non-DL methods in most studies.

CONCLUSION

The DL-based imputation models are a family of techniques, with diverse network structures. Their designation in healthcare is usually tailored to data types with different characteristics. Although DL-based imputation models may not be superior to conventional approaches across all datasets, it is highly possible for them to achieve satisfactory results for a particular data type or dataset. There are, however, still issues with regard to portability, interpretability, and fairness associated with current DL-based imputation models.

Combining Radiological and Genomic TB Portals Data for Drug Resistance Analysis

Tuberculosis (TB) drug resistance is a worldwide public health problem. It decreases the likelihood of a positive outcome for the individual patient and increases the likelihood of disease spread. Therefore, early detection of TB drug resistance is crucial for improving outcomes and controlling disease transmission. While drug-sensitive tuberculosis cases are declining worldwide because of effective treatment, the threat of drug-resistant tuberculosis is growing, and the success rate of drug-resistant tuberculosis treatment is only around 60%. The TB Portals program provides a publicly accessible repository of TB case data with an emphasis on collecting drug-resistant cases. The dataset includes multi-modal information such as socioeconomic/geographic data, clinical characteristics, pathogen genomics, and radiological features. The program is an international collaboration whose participants are typically under a substantial burden of drug-resistant tuberculosis, with data collected from standard clinical care provided to the patients. Consequentially, the TB Portals dataset is heterogenous in nature, with data representing multiple treatment centers in different countries and containing cross-domain information. This study presents the challenges and methods used to address them when working with this real-world dataset. Our goal was to evaluate whether combining radiological features derived from a chest X-ray of the host and genomic features from the pathogen can potentially improve the identification of the drug susceptibility type, drug-sensitive (DS-TB) or drug-resistant (DR-TB), and the length of the first successful drug regimen. To perform these studies, significantly imbalanced data needed to be processed, which included a much larger number of DR-TB cases than DS-TB, many more cases with radiological findings than genomic ones, and the sparse high dimensional nature of the genomic information. Three evaluation studies were carried out. First, the DR-TB/DS-TB classification model achieved an average accuracy of 92.4% when using genomic features alone or when combining radiological and genomic features. Second, the regression model for the length of the first successful treatment had a relative error of 53.5% using radiological features, 25.6% using genomic features, and 22.0% using both radiological and genomic features. Finally, the relative error of the third regression model predicting the length of the first treatment using the most common drug combination varied depending on the feature type used. When using radiological features alone, the relative error was 17.8%. For genomic features alone, the relative error increased to 19.9%. The model had a relative error of 19.0% when both radiological and genomic features were combined. Although combining radiological and genomic features did not improve upon the use of genomic features when classifying DR-TB/DS-TB, the combination of the two feature types improved the relative error of the predictive model for the length of the first successful treatment. Furthermore, the regression model trained on radiological features achieved the best performance when predicting the treatment length of the most common drug combination.

WebPPG: Feasibility and Usability of Self-Performed, Browser-Based Smartphone Photoplethysmography

Smartphones enable and facilitate biomedical studies as they allow the recording of various biomedical signals, including photoplethysmograms (PPG). However, user engagement rates in mobile health studies are reduced when an application (app) needs to be installed. This could be alleviated by using installation-free web apps. We evaluate the feasibility of browser-based PPG recording, conducting the first usability study on smartphone-based PPG. We present an at-home study using a web app and library for PPG recording using the rear camera and flash. The underlying library is freely made available to researchers. 25 Android users participated, using their own smartphones. The study consisted of a demographic and anamnestic questionnaire, the signal recording itself (60 s), and a consecutive usability questionnaire. After filtering, heart rate was extracted (14/17 successful), signal-to-noise ratios assessed (0.64 ± 0.50 dB, mean ± standard deviation), and quality was visually inspected (12/17 usable for diagnosis). Recording was not supported in 9 cases. This was due to the browser's insufficient support for the flash light API. The app received a System Usability Scale score of 82 ± 9, which is above the 90th percentile. Overall, browser flash light support is the main limiting factor for broad device support. Thus, browser-based PPG is not yet widely applicable, although most participants feel comfortable with the recording itself. The utilization of the user-facing camera might represent a more promising approach. This study contributes to the development of low-barrier, user-friendly, installation-free smartphone signal acquisition. This enables profound, comprehensive data collection for research and clinical practice.Clinical relevance- WebPPG offers low-barrier remote diagnostic capabilities without the need for app installation.

An analysis of Chinese nursing electronic medical records to predict violence in psychiatric inpatients using text mining and machine learning techniques

BACKGROUND

The prevalence of violence in acute psychiatric wards is a critical concern. According to a meta-analysis investigating violence in psychiatric inpatient units, researchers estimated that approximately 17% of inpatients commit one or more acts of violence during their stay. Inpatient violence negatively affects health-care providers and patients and may contribute to high staff turnover. Therefore, predicting which psychiatric inpatients will commit violence is of considerable clinical significance.

OBJECTIVE

The present study aimed to estimate the violence rate for psychiatric inpatients and establish a predictive model for violence in psychiatric inpatients.

METHODS

We collected the structured and unstructured data from Chinese nursing electronic medical records (EMRs) for the violence prediction. The data was obtained from the psychiatry department of a regional hospital in southern Taiwan, covering the period between January 2008 and December 2018. Several text mining and machine learning techniques were employed to analyze the data.

RESULTS

The results demonstrated that the rate of violence in psychiatric inpatients is 19.7%. The patients with violence in psychiatric wards were generally younger, had a more violent history, and were more likely to be unmarried. Furthermore, our study supported the feasibility of predicting aggressive incidents in psychiatric wards by using nursing EMRs and the proposed method can be incorporated into routine clinical practice to enable early prediction of inpatient violence.

CONCLUSIONS

Our findings may provide clinicians with a new basis for judgment of the risk of violence in psychiatric wards.

Radiomic applications in upper gastrointestinal cancer surgery

INTRODUCTION

Cross-sectional imaging plays an integral role in the management of upper gastrointestinal (UGI) cancer, from initial diagnosis and staging to determining appropriate treatment strategies. Subjective imaging interpretation has known limitations. The field of radiomics has evolved to extract quantitative data from medical imaging and relate these to biological processes. The key concept behind radiomics is that the high-throughput analysis of quantitative imaging features can provide predictive or prognostic information, with the goal of providing individualised care.

OBJECTIVE

Radiomic studies have shown promising utility in upper gastrointestinal oncology, highlighting a potential role in determining stage of disease and degree of tumour differentiation and predicting recurrence-free survival. This narrative review aims to provide an insight into the concepts underpinning radiomics, as well as its potential applications for guiding treatment and surgical decision-making in upper gastrointestinal malignancy.

CONCLUSION

Outcomes from studies to date have been promising; however, further standardisation and collaboration are required. Large prospective studies with external validation and evaluation of radiomic integration into clinical pathways are needed. Future research should now focus on translating the promising utility of radiomics into meaningful patient outcomes.

Analysis of a National Database Investigating Development of Trigger Finger after Treatment of Dupuytren Disease

Dupuytren disease is associated with inflammation and myofibroblast overgrowth, as is stenosing tenosynovitis (trigger finger). Both are linked with fibroblast proliferation, but a potential associative link between the diseases is unknown. The purpose of this study was to evaluate the progression of trigger finger following treatment for Dupuytren contracture in a large database.

Methods

A commercial database encompassing 53 million patients was utilized from January 1, 2010 to March 31, 2020. The study cohort included patients diagnosed with either Dupuytren disease or trigger finger utilizing International Classification Codes 9 and 10. Terminology codes were used to identify common Dupuytren procedures, as well as trigger finger release. Logistic regression analysis was used to define independent risk factors for developing trigger finger.

Results

A total of 593,606 patients were diagnosed with trigger finger. Of these patients, 15,416 (2.6%) were diagnosed with trigger finger after diagnosis of Dupuytren disease, whereas 2603 (0.4%) patients were diagnosed with trigger finger after treatment of Dupuytren contracture. Independent risk factors for trigger finger included age 65 years or older (OR 1.00, P < 0.05), diabetes (OR 1.12, P < 0.05) and obesity (OR 1.20, P < 0.005). Patients who received collagenase clostridium histolyticum treatment (OR 0.34, P < 0.005) for Dupuytren contracture were significantly less likely to develop trigger finger.

Conclusions

Dupuytren contracture is associated with inflammation and subsequent trigger finger development at a higher rate than the background population frequency. Collagenase clostridium histolyticum injection may lead to a decreased risk of trigger finger requiring surgical intervention in patients with risk factors.

Epidemiology and factors associated with cannabis use among patients with glaucoma in the All of Us Research Program

Purpose

To examine the epidemiology and factors of cannabis use among open-angle glaucoma (OAG) patients.

Methods

In this cross-sectional study, OAG participants in the All of Us database were included. Cannabis ever-users were defined based on record of cannabis use. Demographic and socioeconomic data were collected and compared between cannabis ever-users and never-users using Chi-Square tests and logistic regression. Odds ratios (OR) of potential factors associated with cannabis use were examined in univariable and multivariable models.

Results

Among 3723 OAG participants, 1436 (39%) were cannabis ever-users. The mean (SD) age of never-users and ever-users was 72.9 (10.4) and 69.2 (9.6) years, respectively (P < 0.001). Compared to never-users, Black (34%) and male (55%) participants were better represented in ever-users, while Hispanic or Latino participants (6%) were less represented (P < 0.001). Diversity was also observed in socioeconomic characteristics including marital status, housing security, and income/education levels. A higher percentage of ever-users had a degree ≥12 grades (91%), salaried employment (26%), housing insecurity (12%), and history of cigar smoking (48%), alcohol consumption (96%), and other substance use (47%) (P < 0.001). In the multivariable analysis, Black race (OR [95% CI] = 1.33 [1.06, 1.68]), higher education (OR = 1.19 [1.07, 1.32]), and history of nicotine product smoking (OR: 2.04-2.83), other substance use (OR = 8.14 [6.63, 10.04]), and alcohol consumption (OR = 6.80 [4.45, 10.79]) were significant factors associated with cannabis use. Increased age (OR = 0.96 [0.95, 0.97]), Asian race (OR = 0.18 [0.09, 0.33]), and Hispanic/Latino ethnicity (OR = 0.43 [0.27, 0.68]) were associated with decreased odds of use (P < 0.02).

Conclusions

This study elucidated the previously uncharacterized epidemiology and factors associated with cannabis use among OAG patients, which may help to identify patients requiring additional outreach on unsupervised marijuana use.

Explainable discovery of disease biomarkers: The case of ovarian cancer to illustrate the best practice in machine learning and Shapley analysis

OBJECTIVE

Ovarian cancer is a significant health issue with lasting impacts on the community. Despite recent advances in surgical, chemotherapeutic and radiotherapeutic interventions, they have had only marginal impacts due to an inability to identify biomarkers at an early stage. Biomarker discovery is challenging, yet essential for improving drug discovery and clinical care. Machine learning (ML) techniques are invaluable for recognising complex patterns in biomarkers compared to conventional methods, yet they can lack physical insights into diagnosis. eXplainable Artificial Intelligence (XAI) is capable of providing deeper insights into the decision-making of complex ML algorithms increasing their applicability. We aim to introduce best practice for combining ML and XAI techniques for biomarker validation tasks.

METHODS

We focused on classification tasks and a game theoretic approach based on Shapley values to build and evaluate models and visualise results. We described the workflow and apply the pipeline in a case study using the CDAS PLCO Ovarian Biomarkers dataset to demonstrate the potential for accuracy and utility.

RESULTS

The case study results demonstrate the efficacy of the ML pipeline, its consistency, and advantages compared to conventional statistical approaches.

CONCLUSION

The resulting guidelines provide a general framework for practical application of XAI in medical research that can inform clinicians and validate and explain cancer biomarkers.

Post-Stroke Infections: Insights from Big Data Using Clinical Data Warehouse (CDW)

This study analyzed a digitized database of electronic medical records (EMRs) to identify risk factors for post-stroke infections. The sample included 41,236 patients hospitalized with a first stroke diagnosis (ICD-10 codes I60, I61, I63, and I64) between January 2011 and December 2020. Logistic regression analysis was performed to examine the effect of clinical variables on post-stroke infection. Multivariable analysis revealed that post-stroke infection was associated with the male sex (odds ratio [OR]: 1.79; 95% confidence interval [CI]: 1.49-2.15), brain surgery (OR: 7.89; 95% CI: 6.27-9.92), mechanical ventilation (OR: 18.26; 95% CI: 8.49-44.32), enteral tube feeding (OR: 3.65; 95% CI: 2.98-4.47), and functional activity level (modified Barthel index: OR: 0.98; 95% CI: 0.98-0.98). In addition, exposure to steroids (OR: 2.22; 95% CI: 1.60-3.06) and acid-suppressant drugs (OR: 1.44; 95% CI: 1.15-1.81) increased the risk of infection. On the basis of the findings from this multicenter study, it is crucial to carefully evaluate the balance between the potential benefits of acid-suppressant drugs or corticosteroids and the increased risk of infection in patients at high risk for post-stroke infection.

Machine learning of electrophysiological signals for the prediction of ventricular arrhythmias: systematic review and examination of heterogeneity between studies

BACKGROUND

Ventricular arrhythmia (VA) precipitating sudden cardiac arrest (SCD) is among the most frequent causes of death and pose a high burden on public health systems worldwide. The increasing availability of electrophysiological signals collected through conventional methods (e.g. electrocardiography (ECG)) and digital health technologies (e.g. wearable devices) in combination with novel predictive analytics using machine learning (ML) and deep learning (DL) hold potential for personalised predictions of arrhythmic events.

METHODS

This systematic review and exploratory meta-analysis assesses the state-of-the-art of ML/DL models of electrophysiological signals for personalised prediction of malignant VA or SCD, and studies potential causes of bias (PROSPERO, reference: CRD42021283464). Five electronic databases were searched to identify eligible studies. Pooled estimates of the diagnostic odds ratio (DOR) and summary area under the curve (AUROC) were calculated. Meta-analyses were performed separately for studies using publicly available, ad-hoc datasets, versus targeted clinical data acquisition. Studies were scored on risk of bias by the PROBAST tool.

FINDINGS

2194 studies were identified of which 46 were included in the systematic review and 32 in the meta-analysis. Pooling of individual models demonstrated a summary AUROC of 0.856 (95% CI 0.755-0.909) for short-term (time-to-event up to 72 h) prediction and AUROC of 0.876 (95% CI 0.642-0.980) for long-term prediction (time-to-event up to years). While models developed on ad-hoc sets had higher pooled performance (AUROC 0.919, 95% CI 0.867-0.952), they had a high risk of bias related to the re-use and overlap of small ad-hoc datasets, choices of ML tool and a lack of external model validation.

INTERPRETATION

ML and DL models appear to accurately predict malignant VA and SCD. However, wide heterogeneity between studies, in part due to small ad-hoc datasets and choice of ML model, may reduce the ability to generalise and should be addressed in future studies.

FUNDING

This publication is part of the project DEEP RISK ICD (with project number 452019308) of the research programme Rubicon which is (partly) financed by the Dutch Research Council (NWO). This research is partly funded by the Amsterdam Cardiovascular Sciences (personal grant F.V.Y.T).

Recent Studies of Artificial Intelligence on In Silico Drug Distribution Prediction

Drug distribution is an important process in pharmacokinetics because it has the potential to influence both the amount of medicine reaching the active sites and the effectiveness as well as safety of the drug. The main causes of 90% of drug failures in clinical development are lack of efficacy and uncontrolled toxicity. In recent years, several advances and promising developments in drug distribution property prediction have been achieved, especially in silico, which helped to drastically reduce the time and expense of screening undesired drug candidates. In this study, we provide comprehensive knowledge of drug distribution background, influencing factors, and artificial intelligence-based distribution property prediction models from 2019 to the present. Additionally, we gathered and analyzed public databases and datasets commonly utilized by the scientific community for distribution prediction. The distribution property prediction performance of five large ADMET prediction tools is mentioned as a benchmark for future research. On this basis, we also offer future challenges in drug distribution prediction and research directions. We hope that this review will provide researchers with helpful insight into distribution prediction, thus facilitating the development of innovative approaches for drug discovery.