Trends in biomedical informatics: automated topic analysis of JAMIA articles

Disparities of Health Program Information Systems in Indonesia: A Cross-Sectional Indonesian Health Facility Research 2019

Although a recording and reporting format for health centers already exists for Indonesia's standard information system, numerous health applications still need to meet the needs of each program. Therefore, this study aimed to demonstrate the potential disparities in information systems in the application and data collection of health programs among Indonesian community health centers (CHCs) based on provinces and regions. This cross-sectional research used data from 9831 CHCs from the Health Facilities Research 2019 (RIFASKES). Significance was assessed using a chi-square test and analysis of variance (ANOVA). The number of applications was depicted on a map using the spmap command with STATA version 14. It showed that region 2, which represented Java and Bali, was the best, followed by regions 1, which comprised Sumatra Island and its surroundings, and 3, Nusa Tenggara. The highest mean, equaling that of Java, was discovered in three provinces of region 1, namely, Jambi, Lampung, and Bangka Belitung. Furthermore, Papua and West Papua had less than 60% for all types of data-storage programs. Hence, there is a disparity in the health information system in Indonesia by province and region. The results of this analysis recommend future improvement of the CHCs' information systems.

Machine Learning Approaches to Predict Risks of Diabetic Complications and Poor Glycemic Control in Nonadherent Type 2 Diabetes

Purpose: The objective of this study was to evaluate the efficacy of machine learning algorithms in predicting risks of complications and poor glycemic control in nonadherent type 2 diabetes (T2D). Materials and Methods: This study was a real-world study of the complications and blood glucose prognosis of nonadherent T2D patients. Data of inpatients in Sichuan Provincial People's Hospital from January 2010 to December 2015 were collected. The T2D patients who had neither been monitored for glycosylated hemoglobin A nor had changed their hyperglycemia treatment regimens within the last 12 months were the object of this study. Seven types of machine learning algorithms were used to develop 18 prediction models. The predictive performance was mainly assessed using the area under the curve of the testing set. Results: Of 800 T2D patients, 165 (20.6%) met the inclusion criteria, of which 129 (78.2%) had poor glycemic control (defined as glycosylated hemoglobin A ≥7%). The highest area under the curves of the testing set for diabetic nephropathy, diabetic peripheral neuropathy, diabetic angiopathy, diabetic eye disease, and glycosylated hemoglobin A were 0.902 ± 0.040, 0.859 ± 0.050, 0.889 ± 0.059, 0.832 ± 0.086, and 0.825 ± 0.092, respectively. Conclusion: Both univariate analysis and machine learning methods reached the same conclusion. The duration of T2D and the duration of unadjusted hypoglycemic treatment were the key risk factors of diabetic complications, and the number of hypoglycemic drugs was the key risk factor of glycemic control of nonadherent T2D. This was the first study to use machine learning algorithms to explore the potential adverse outcomes of nonadherent T2D. The performances of the final prediction models we developed were acceptable; our prediction performances outperformed most other previous studies in most evaluation measures. Those models have potential clinical applicability in improving T2D care.

A Review of Statistical and Machine Learning Techniques for Microvascular Complications in Type 2 Diabetes

Background and Introduction: Diabetes mellitus is a metabolic disorder that has emerged as a serious public health issue worldwide. According to the World Health Organization (WHO), without interventions, the number of diabetic incidences is expected to be at least 629 million by 2045. Uncontrolled diabetes gradually leads to progressive damage to eyes, heart, kidneys, blood vessels, and nerves.

METHODS

The paper presents a critical review of existing statistical and Artificial Intelligence (AI) based machine learning techniques with respect to DM complications, mainly retinopathy, neuropathy, and nephropathy. The statistical and machine learning analytic techniques are used to structure the subsequent content review.

RESULTS

It has been observed that statistical analysis can help only in inferential and descriptive analysis whereas, AI-based machine learning models can even provide actionable prediction models for faster and accurate diagnosis of complications associated with DM.

CONCLUSION

The integration of AI-based analytics techniques, like machine learning and deep learning in clinical medicine, will result in improved disease management through faster disease detection and cost reduction for the treatment.

Topic Analysis of UK Fitness to Practise Cases: What Lessons Can Be Learnt?

Background: Fitness to practise (FtP) impairment (failure of a healthcare professional to demonstrate skills, knowledge, character and/or health required for their job) can compromise patient safety, the profession's reputation, and an individual's career. In the United Kingdom (UK), various healthcare professionals' FtP cases (documents about the panel hearing(s) and outcome(s) relating to the alleged FtP impairment) are publicly available, yet reviewing these to learn lessons may be time-consuming given the number of cases across the professions and amount of text in each. We aimed to demonstrate how machine learning facilitated the examination of such cases (at uni- and multi-professional level), involving UK dental, medical, nursing and pharmacy professionals. Methods: Cases dating from August 2017 to June 2019 were downloaded (577 dental, 481 medical, 2199 nursing and 63 pharmacy) and converted to text files. A topic analysis method (non-negative matrix factorization; machine learning) was employed for data analysis. Results: Identified topics were criminal offences; dishonesty (fraud and theft); drug possession/supply; English language; indemnity insurance; patient care (including incompetence) and personal behavior (aggression, sexual conduct and substance misuse). The most frequently identified topic for dental, medical and nursing professions was patient care whereas for pharmacy, it was criminal offences. Conclusions: While commonalities exist, each has different priorities which professional and educational organizations should strive to address.

Multi-objective Parameter Auto-tuning for Tissue Image Segmentation Workflows

We propose a software platform that integrates methods and tools for multi-objective parameter auto-tuning in tissue image segmentation workflows. The goal of our work is to provide an approach for improving the accuracy of nucleus/cell segmentation pipelines by tuning their input parameters. The shape, size, and texture features of nuclei in tissue are important biomarkers for disease prognosis, and accurate computation of these features depends on accurate delineation of boundaries of nuclei. Input parameters in many nucleus segmentation workflows affect segmentation accuracy and have to be tuned for optimal performance. This is a time-consuming and computationally expensive process; automating this step facilitates more robust image segmentation workflows and enables more efficient application of image analysis in large image datasets. Our software platform adjusts the parameters of a nuclear segmentation algorithm to maximize the quality of image segmentation results while minimizing the execution time. It implements several optimization methods to search the parameter space efficiently. In addition, the methodology is developed to execute on high-performance computing systems to reduce the execution time of the parameter tuning phase. These capabilities are packaged in a Docker container for easy deployment and can be used through a friendly interface extension in 3D Slicer. Our results using three real-world image segmentation workflows demonstrate that the proposed solution is able to (1) search a small fraction (about 100 points) of the parameter space, which contains billions to trillions of points, and improve the quality of segmentation output by × 1.20, × 1.29, and × 1.29, on average; (2) decrease the execution time of a segmentation workflow by up to 11.79× while improving output quality; and (3) effectively use parallel systems to accelerate parameter tuning and segmentation phases.

Evaluating the outcomes of medical informatics development as a discipline in China: A publication perspective

BACKGROUND

As the world's second largest economy, China makes unique contributions to the world in many fields, including sociology, the economy, technology and defense. Medical informatics (MI) is an important cross-disciplinary field that, along with its applications, has received massive funding from the Chinese government. However, the question of how to evaluate China's input and output in MI remains important and complex issue of great significance for China and the rest of the world.

OBJECTIVE

This paper analyzed, for the first time, the quality and quantity of research by Chinese academics in MI based on their articles published in international specialty journals in recent years and examined MI research hotspots in China. Our purpose is to summarize the experiences and lessons learned by China and the rest of the world as they develop MI.

METHOD

We targeted 18 MI journals listed in the JCR 2016 report and searched for papers published by Chinese academics in these 18 journals in the WOS and PUBMED databases and on journal sites. We also performed data cleansing and categorized the obtained information. We used Excel, SPSS, Ucinet and NetDraw to conduct quantitative analyses on the research papers.

RESULTS

A total of 1340 articles satisfied the inclusion criteria of this study. We observed a significant upward trend in the number of articles published over time, particularly after 2011. Lei Jianbo, Huang Zhengxing and Li Jin-song are active Chinese authors in the MI discipline who have written many high-quality publications. Meanwhile, universities remain the primary breeding grounds for scientific research: 93.36% of the articles came from universities. Zhejiang University published the most first-author articles, whereas Zhejiang University, Shanghai Jiao Tong University and Tsinghua University produced 17.76% of the total articles. According to the lists of authors, 24% of the papers were co-authored with foreign researchers. This rate of cooperation is increasing each year, from 5.88% to the current rate of 39.04%. An analysis of keywords showed that "EMR", "SVM", "Authentication", "Telecare medical information system", "EEG", "ECG" and "RFID" were the most frequently searched terms in popular technological fields, including artificial intelligence and image processing. In recent years, there has been a significant increase in the number of high-frequency keywords and a broadening range of research fields, which has led to the emergence of several research hotspots, including MI systems, mobile health care, telecare, data mining and machine learning.

CONCLUSIONS

Through the quantitative analysis of publications, we discovered the emergence of three stages - infancy, slow growth and rapid growth - in China's MI research in recent years as academics make achievements in their research works. The global influence of Chinese academics is growing, and they are making increasingly conscious efforts to enter into research collaborations with foreign researchers. The findings of Chinese academics' publications are gaining international recognition.

Developing Embedded Taxonomy and Mining Patients' Interests From Web-Based Physician Reviews: Mixed-Methods Approach

BACKGROUND

Web-based physician reviews are invaluable gold mines that merit further investigation. Although many studies have explored the text information of physician reviews, very few have focused on developing a systematic topic taxonomy embedded in physician reviews. The first step toward mining physician reviews is to determine how the natural structure or dimensions is embedded in reviews. Therefore, it is relevant to develop the topic taxonomy rigorously and systematically.

OBJECTIVE

This study aims to develop a hierarchical topic taxonomy to uncover the latent structure of physician reviews and illustrate its application for mining patients' interests based on the proposed taxonomy and algorithm.

METHODS

Data comprised 122,716 physician reviews, including reviews of 8501 doctors from a leading physician review website in China (haodf.com), collected between 2007 and 2015. Mixed methods, including a literature review, data-driven-based topic discovery, and human annotation were used to develop the physician review topic taxonomy.

RESULTS

The identified taxonomy included 3 domains or high-level categories and 9 subtopics or low-level categories. The physician-related domain included the categories of medical ethics, medical competence, communication skills, medical advice, and prescriptions. The patient-related domain included the categories of the patient profile, symptoms, diagnosis, and pathogenesis. The system-related domain included the categories of financing and operation process. The F-measure of the proposed classification algorithm reached 0.816 on average. Symptoms (Cohen d=1.58, Δu=0.216, t=229.75, and P<.001) are more often mentioned by patients with acute diseases, whereas communication skills (Cohen d=-0.29, Δu=-0.038, t=-42.01, and P<.001), financing (Cohen d=-0.68, Δu=-0.098, t=-99.26, and P<.001), and diagnosis and pathogenesis (Cohen d=-0.55, Δu=-0.078, t=-80.09, and P<.001) are more often mentioned by patients with chronic diseases. Patients with mild diseases were more interested in medical ethics (Cohen d=0.25, Δu 0.039, t=8.33, and P<.001), operation process (Cohen d=0.57, Δu 0.060, t=18.75, and P<.001), patient profile (Cohen d=1.19, Δu 0.132, t=39.33, and P<.001), and symptoms (Cohen d=1.91, Δu=0.274, t=62.82, and P<.001). Meanwhile, patients with serious diseases were more interested in medical competence (Cohen d=-0.99, Δu=-0.165, t=-32.58, and P<.001), medical advice and prescription (Cohen d=-0.65, Δu=-0.082, t=-21.45, and P<.001), financing (Cohen d=-0.26, Δu=-0.018, t=-8.45, and P<.001), and diagnosis and pathogenesis (Cohen d=-1.55, Δu=-0.229, t=-50.93, and P<.001).

CONCLUSIONS

This mixed-methods approach, integrating literature reviews, data-driven topic discovery, and human annotation, is an effective and rigorous way to develop a physician review topic taxonomy. The proposed algorithm based on Labeled-Latent Dirichlet Allocation can achieve impressive classification results for mining patients' interests. Furthermore, the mining results reveal marked differences in patients' interests across different disease types, socioeconomic development levels, and hospital levels.

Machine Learning Methods to Predict Diabetes Complications

One of the areas where Artificial Intelligence is having more impact is machine learning, which develops algorithms able to learn patterns and decision rules from data. Machine learning algorithms have been embedded into data mining pipelines, which can combine them with classical statistical strategies, to extract knowledge from data. Within the EU-funded MOSAIC project, a data mining pipeline has been used to derive a set of predictive models of type 2 diabetes mellitus (T2DM) complications based on electronic health record data of nearly one thousand patients. Such pipeline comprises clinical center profiling, predictive model targeting, predictive model construction and model validation. After having dealt with missing data by means of random forest (RF) and having applied suitable strategies to handle class imbalance, we have used Logistic Regression with stepwise feature selection to predict the onset of retinopathy, neuropathy, or nephropathy, at different time scenarios, at 3, 5, and 7 years from the first visit at the Hospital Center for Diabetes (not from the diagnosis). Considered variables are gender, age, time from diagnosis, body mass index (BMI), glycated hemoglobin (HbA1c), hypertension, and smoking habit. Final models, tailored in accordance with the complications, provided an accuracy up to 0.838. Different variables were selected for each complication and time scenario, leading to specialized models easy to translate to the clinical practice.

Tensor Factorization for Precision Medicine in Heart Failure with Preserved Ejection Fraction

Heart failure with preserved ejection fraction (HFpEF) is a heterogeneous clinical syndrome that may benefit from improved subtyping in order to better characterize its pathophysiology and to develop novel targeted therapies. The United States Precision Medicine Initiative comes amid the rapid growth in quantity and modality of clinical data for HFpEF patients ranging from deep phenotypic to trans-omic data. Tensor factorization, a form of machine learning, allows for the integration of multiple data modalities to derive clinically relevant HFpEF subtypes that may have significant differences in underlying pathophysiology and differential response to therapies. Tensor factorization also allows for better interpretability by supporting dimensionality reduction and identifying latent groups of data for meaningful summarization of both features and disease outcomes. In this narrative review, we analyze the modest literature on the application of tensor factorization to related biomedical fields including genotyping and phenotyping. Based on the cited work including work of our own, we suggest multiple tensor factorization formulations capable of integrating the deep phenotypic and trans-omic modalities of data for HFpEF, or accounting for interactions between genetic variants at different omic hierarchies. We encourage extensive experimental studies to tackle challenges in applying tensor factorization for precision medicine in HFpEF, including effectively incorporating existing medical knowledge, properly accounting for uncertainty, and efficiently enforcing sparsity for better interpretability.

Automated Diabetes Case Identification Using Electronic Health Record Data at a Tertiary Care Facility

Objective

To develop and validate a phenotyping algorithm for the identification of patients with type 1 and type 2 diabetes mellitus (DM) preoperatively using routinely available clinical data from electronic health records.

Patients and Methods

We used first-order logic rules (if-then-else rules) to imply the presence or absence of DM types 1 and 2. The “if” clause of each rule is a conjunction of logical and, or predicates that provides evidence toward or against the presence of DM. The rule includes International Classification of Diseases, Ninth Revision, Clinical Modification diagnostic codes, outpatient prescription information, laboratory values, and positive annotation of DM in patients’ clinical notes. This study was conducted from March 2, 2015, through February 10, 2016. The performance of our rule-based approach and similar approaches proposed by other institutions was evaluated with a reference standard created by an expert reviewer and implemented for routine clinical care at an academic medical center.

Results

A total of 4208 surgical patients (mean age, 52 years; males, 48%) were analyzed to develop the phenotyping algorithm. Expert review identified 685 patients (16.28% of the full cohort) as having DM. Our proposed method identified 684 patients (16.25%) as having DM. The algorithm performed well—99.70% sensitivity, 99.97% specificity—and compared favorably with previous approaches.

Conclusion

Among patients undergoing surgery, determination of DM can be made with high accuracy using simple, computationally efficient rules. Knowledge of patients’ DM status before surgery may alter physicians’ care plan and reduce postsurgical complications. Nevertheless, future efforts are necessary to determine the effect of first-order logic rules on clinical processes and patient outcomes.

General guidelines for biomedical software development

Most bioinformatics tools available today were not written by professional software developers, but by people that wanted to solve their own problems, using computational solutions and spending the minimum time and effort possible, since these were just the means to an end. Consequently, a vast number of software applications are currently available, hindering the task of identifying the utility and quality of each. At the same time, this situation has hindered regular adoption of these tools in clinical practice. Typically, they are not sufficiently developed to be used by most clinical researchers and practitioners. To address these issues, it is necessary to re-think how biomedical applications are built and adopt new strategies that ensure quality, efficiency, robustness, correctness and reusability of software components. We also need to engage end-users during the development process to ensure that applications fit their needs. In this review, we present a set of guidelines to support biomedical software development, with an explanation of how they can be implemented and what kind of open-source tools can be used for each specific topic.