Predicting Hepatocellular Carcinoma With Minimal Features From Electronic Health Records: Development of a Deep Learning Model.

BACKGROUND: Hepatocellular carcinoma (HCC), usually known as hepatoma, is the third leading cause of cancer mortality globally. Early detection of HCC helps in its treatment and increases survival rates. OBJECTIVE: The aim of this study is to develop a deep learning model, using the trend and severity of each medical event from the electronic health record to accurately predict the patients who will be diagnosed with HCC in 1 year. METHODS: Patients with HCC were screened out from the National Health Insurance Research Database of Taiwan between 1999 and 2013. To be included, the patients with HCC had to register as patients with cancer in the catastrophic illness file and had to be diagnosed as a patient with HCC in an inpatient admission. The control cases (non-HCC patients) were randomly sampled from the same database. We used age, gender, diagnosis code, drug code, and time information as the input variables of a convolution neural network model to predict those patients with HCC. We also inspected the highly weighted variables in the model and compared them to their odds ratio at HCC to understand how the predictive model works. RESULTS: We included 47,945 individuals, 9553 of whom were patients with HCC. The area under the receiver operating curve (AUROC) of the model for predicting HCC risk 1 year in advance was 0.94 (95% CI 0.937-0.943), with a sensitivity of 0.869 and a specificity 0.865. The AUROC for predicting HCC patients 7 days, 6 months, 1 year, 2 years, and 3 years early were 0.96, 0.94, 0.94, 0.91, and 0.91, respectively. CONCLUSIONS: The findings of this study show that the convolutional neural network model has immense potential to predict the risk of HCC 1 year in advance with minimal features available in the electronic health records.

Development of a Web-Based System for Exploring Cancer Risk With Long-term Use of Drugs: Logistic Regression Approach.

BACKGROUND: Existing epidemiological evidence regarding the association between the long-term use of drugs and cancer risk remains controversial. OBJECTIVE: We aimed to have a comprehensive view of the cancer risk of the long-term use of drugs. METHODS: A nationwide population-based, nested, case-control study was conducted within the National Health Insurance Research Database sample cohort of 1999 to 2013 in Taiwan. We identified cases in adults aged 20 years and older who were receiving treatment for at least two months before the index date. We randomly selected control patients from the patients without a cancer diagnosis during the 15 years (1999-2013) of the study period. Case and control patients were matched 1:4 based on age, sex, and visit date. Conditional logistic regression was used to estimate the association between drug exposure and cancer risk by adjusting potential confounders such as drugs and comorbidities. RESULTS: There were 79,245 cancer cases and 316,980 matched controls included in this study. Of the 45,368 associations, there were 2419, 1302, 662, and 366 associations found statistically significant at a level of P<.05, P<.01, P<.001, and P<.0001, respectively. Benzodiazepine derivatives were associated with an increased risk of brain cancer (adjusted odds ratio [AOR] 1.379, 95% CI 1.138-1.670; P=.001). Statins were associated with a reduced risk of liver cancer (AOR 0.470, 95% CI 0.426-0.517; P<.0001) and gastric cancer (AOR 0.781, 95% CI 0.678-0.900; P<.001). Our web-based system, which collected comprehensive data of associations, contained 2 domains: (1) the drug and cancer association page and (2) the overview page. CONCLUSIONS: Our web-based system provides an overview of comprehensive quantified data of drug-cancer associations. With all the quantified data visualized, the system is expected to facilitate further research on cancer risk and prevention, potentially serving as a stepping-stone to consulting and exploring associations between the long-term use of drugs and cancer risk.

Machine Learning Approach to Reduce Alert Fatigue Using a Disease Medication-Related Clinical Decision Support System: Model Development and Validation.

BACKGROUND: Computerized physician order entry (CPOE) systems are incorporated into clinical decision support systems (CDSSs) to reduce medication errors and improve patient safety. Automatic alerts generated from CDSSs can directly assist physicians in making useful clinical decisions and can help shape prescribing behavior. Multiple studies reported that approximately 90%-96% of alerts are overridden by physicians, which raises questions about the effectiveness of CDSSs. There is intense interest in developing sophisticated methods to combat alert fatigue, but there is no consensus on the optimal approaches so far. OBJECTIVE: Our objective was to develop machine learning prediction models to predict physicians' responses in order to reduce alert fatigue from disease medication-related CDSSs. METHODS: We collected data from a disease medication-related CDSS from a university teaching hospital in Taiwan. We considered prescriptions that triggered alerts in the CDSS between August 2018 and May 2019. Machine learning models, such as artificial neural network (ANN), random forest (RF), naïve Bayes (NB), gradient boosting (GB), and support vector machine (SVM), were used to develop prediction models. The data were randomly split into training (80%) and testing (20%) datasets. RESULTS: A total of 6453 prescriptions were used in our model. The ANN machine learning prediction model demonstrated excellent discrimination (area under the receiver operating characteristic curve [AUROC] 0.94; accuracy 0.85), whereas the RF, NB, GB, and SVM models had AUROCs of 0.93, 0.91, 0.91, and 0.80, respectively. The sensitivity and specificity of the ANN model were 0.87 and 0.83, respectively. CONCLUSIONS: In this study, ANN showed substantially better performance in predicting individual physician responses to an alert from a disease medication-related CDSS, as compared to the other models. To our knowledge, this is the first study to use machine learning models to predict physician responses to alerts; furthermore, it can help to develop sophisticated CDSSs in real-world clinical settings.

Application of a Common Data Model (CDM) to rank the paediatric user and prescription prevalence of 15 different drug classes in South Korea, Hong Kong, Taiwan, Japan and Australia: an observational, descriptive study.

OBJECTIVE: To measure the paediatric user and prescription prevalence in inpatient and ambulatory settings in South Korea, Hong Kong, Taiwan, Japan and Australia by age and gender. A further objective was to list the most commonly used drugs per drug class, per country. DESIGN AND SETTING: Hospital inpatient and insurance paediatric healthcare data from the following databases were used to conduct this descriptive drug utilisation study: (i) the South Korean Ajou University School of Medicine database; (ii) the Hong Kong Clinical Data Analysis and Reporting System; (iii) the Japan Medical Data Center; (iv) Taiwan's National Health Insurance Research Database and (v) the Australian Pharmaceutical Benefits Scheme. Country-specific data were transformed into the Observational Medical Outcomes Partnership Common Data Model. PATIENTS: Children (≤18 years) with at least 1 day of observation in any of the respective databases from January 2009 until December 2013 were included. MAIN OUTCOME MEASURES: For each drug class, we assessed the per-protocol overall user and prescription prevalence rates (per 1000 persons) per country and setting. RESULTS: Our study population comprised 1 574 524 children (52.9% male). The highest proportion of dispensings was recorded in the youngest age category (<2 years) for inpatients (45.1%) with a relatively high user prevalence of analgesics and antibiotics. Adrenergics, antihistamines, mucolytics and corticosteroids were used in 10%-15% of patients. For ambulatory patients, the highest proportion of dispensings was recorded in the middle age category (2-11 years, 67.1%) with antibiotics the most dispensed drug overall. CONCLUSIONS: Country-specific paediatric drug utilisation patterns were described, ranked and compared between four East Asian countries and Australia. The widespread use of mucolytics in East Asia warrants further investigation.

An artificial intelligence approach to early predict non-ST-elevation myocardial infarction patients with chest pain.

BACKGROUND AND AIMS: Hospital admission rate for the patients with chest pain has already been increased worldwide but no existing risk score has been designed to stratify non-ST-elevation myocardial infarction (NSTEMI) from non-cardiogenic chest pain. Clinical diagnosis of chest pain in the emergency department is always highly subjective and variable. We, therefore, aimed to develop an artificial intelligence approach to predict stable NSTEMI that would give valuable insight to reduce misdiagnosis in the real clinical setting. METHODS: A standard protocol was developed to collect data from chest pain patients who had visited the emergency department between December 2016 and February 2017. All the chest pain patients with aged <20 years were primarily included in this study. However, STEMI, previous history of ACS, and out-of-hospital cardiac arrest were excluded from our study. An artificial neural network (ANN) model was then developed to predict NSTEMI patients. The accuracy, sensitivity, specificity, and receiver operating characteristic curve was used to measure the performance of this model. RESULTS: A total of 268 chest pain patients were included in this study; of those, 47 (17.5%) was stable NSTEMI, and 221 (82.5%) was unstable angina patients. Serval risk factors such as cardiac risk factor, systolic blood pressure, hemoglobin, corrected QT interval (QTc), PR interval, glutamic-oxaloacetic transaminase, glutamic pyruvic transaminase and troponin were independently associated with stable NSTEMI. The area under the receiver operating characteristic (AUROC) and accuracy of ANN were 98.4, and 92.86. Additionally, the sensitivity, specificity, positive predictive value, and negative predictive value of the ANN model was 90.91, 93.33, 76.92, and 97.67 respectively. CONCLUSION: Our prediction model showed a higher accuracy to predict NSTEMI patients. This model has a potential application in disease detection, monitoring, and prognosis of chest pain at risk of AMI.

Prediction of fatty liver disease using machine learning algorithms.

BACKGROUND AND OBJECTIVE: Fatty liver disease (FLD) is a common clinical complication; it is associated with high morbidity and mortality. However, an early prediction of FLD patients provides an opportunity to make an appropriate strategy for prevention, early diagnosis and treatment. We aimed to develop a machine learning model to predict FLD that could assist physicians in classifying high-risk patients and make a novel diagnosis, prevent and manage FLD. METHODS: We included all patients who had an initial fatty liver screening at the New Taipei City Hospital between 1st and 31st December 2009. Classification models such as random forest (RF), Naïve Bayes (NB), artificial neural networks (ANN), and logistic regression (LR) were developed to predict FLD. The area under the receiver operating characteristic curve (ROC) was used to evaluate performances among the four models. RESULTS: A total of 577 patients were included in this study; of those 377 patients had fatty liver. The area under the receiver operating characteristic (AUROC) of RF, NB, ANN, and LR with 10 fold-cross validation was 0.925, 0.888, 0.895, and 0.854 respectively. Additionally, The accuracy of RF, NB, ANN, and LR 87.48, 82.65, 81.85, and 76.96%. CONCLUSION: In this study, we developed and compared the four classification models to predict fatty liver disease accurately. However, the random forest model showed higher performance than other classification models. Implementation of a random forest model in the clinical setting could help physicians to stratify fatty liver patients for primary prevention, surveillance, early treatment, and management.

Gout drugs use and risk of cancer: A case-control study.

OBJECTIVE: Firm conclusion about whether short and long-term gout medications use has an impact on cancer risk remain inconclusive. The aim of this study was to investigate the association between gout drugs use and risk of cancer. METHODS: We conducted a retrospective longitudinal population-based case-control study in Taiwan. Cases were identified all patients who were aged 20years or above, and had a first time diagnosis of cancers for the period between 2001 and 2011. Adjusted odds ratios (AORs) and 95% confidence intervals (CIs) were calculated by using conditional logistic regression. RESULTS: We examined 601,733 cases and 2,406,932 matched controls. The adjusted odd ratio for any gout drugs use and overall cancer risk was 1.007 (95% CI: 0.994-1.020). There was a significant risk of leukemia (AOR: 1.34, 95% CI: 1.20-1.50), endometrial cancer (AOR: 1.33, 95% CI: 1.12-1.57), non-Hodgkin's (AOR: 1.24, 95% CI: 1.13-1.35), female breast cancer (AOR: 1.21, 95% CI: 1.13-1.29), cervical cancer (AOR: 1.21, 95% CI: 1.07-1.37). However, no association was observed in male group (AOR: 0.97, 95% CI: 0.95-0.98) but female showed a significantly increased risk of cancer at any site (AOR: 1.107, 95% CI: 1.08-1.13). CONCLUSION: In summary, our results suggest that gout drugs increase risk of the most common cancers, particularly in leukemia, non-Hodgkin's, endometrial, breast and cervical cancer.

Uncovering exposures responsible for birth season - disease effects: a global study.

OBJECTIVE: Birth month and climate impact lifetime disease risk, while the underlying exposures remain largely elusive. We seek to uncover distal risk factors underlying these relationships by probing the relationship between global exposure variance and disease risk variance by birth season. MATERIAL AND METHODS: This study utilizes electronic health record data from 6 sites representing 10.5 million individuals in 3 countries (United States, South Korea, and Taiwan). We obtained birth month-disease risk curves from each site in a case-control manner. Next, we correlated each birth month-disease risk curve with each exposure. A meta-analysis was then performed of correlations across sites. This allowed us to identify the most significant birth month-exposure relationships supported by all 6 sites while adjusting for multiplicity. We also successfully distinguish relative age effects (a cultural effect) from environmental exposures. RESULTS: Attention deficit hyperactivity disorder was the only identified relative age association. Our methods identified several culprit exposures that correspond well with the literature in the field. These include a link between first-trimester exposure to carbon monoxide and increased risk of depressive disorder (R = 0.725, confidence interval [95% CI], 0.529-0.847), first-trimester exposure to fine air particulates and increased risk of atrial fibrillation (R = 0.564, 95% CI, 0.363-0.715), and decreased exposure to sunlight during the third trimester and increased risk of type 2 diabetes mellitus (R = -0.816, 95% CI, -0.5767, -0.929). CONCLUSION: A global study of birth month-disease relationships reveals distal risk factors involved in causal biological pathways that underlie them.

Ankylosing spondylitis and the risk of cancer.

Cancer is a multifactorial disease, and imbalances of the immune response and sex-associated features are considered risk factors for certain types of cancer. The present study aimed to assess whether ankylosing spondylitis (AS), an immune disorder that predominantly affects young adult men, is associated with an increased risk of cancer. Using the Taiwan National Health Insurance Research Database, a cohort of patients diagnosed with AS between 2000 and 2008 who had no history of cancer prior to enrollment was established (n=5,452). Age- and sex-matched patients without AS served as controls (n=21,808). The results revealed that the overall incidence of cancer was elevated in patients with AS [standardized incidence ratio (SIR), 1.15; 95% confidence interval (CI), 1.03-1.27]. AS carried an increased risk of hematological malignancy in both sexes, colon cancer in females and bone and prostate cancer in males. Young patients with AS (≤35 years) and patients with a Charlson comorbidity index (CCI) ≥2 experienced a higher incidence of cancer (males, SIR 1.92, and 95% CI 1.04-3.26; females, SIR 2.00 and 95% CI 1.46-5.50). The cancer risk was increased during the first 3 years following the diagnosis of AS (SIR 1.49, 95% CI 1.29-1.71), and overall cancer-free survival was significantly decreased in patients with AS patients of both sexes (P<0.0001). Therefore, AS was found to be associated with an increased risk of cancer. All AS patients must be screened for hematological malignancies, for prostate and bone cancer in males, and for colon cancer in females, particularly younger patients with a CCI ≥2.

Viral warts (Human Papilloma Virus) as a potential risk for breast cancer among younger females.

INTRODUCTION: There have been several reports on the role of human papillomavirus (HPV) in the etiology of breast cancer. To our knowledge, this is first study to use disease-disease association data-mining approach to analyzing viral warts and breast cancer to be conducted in Taiwanese population. MATERIALS AND METHODS: We analyzed the Taiwan's National Health Insurance database (NHIDM data comprising of 23 million patient data) to examine the association between viral warts and female breast carcinoma. The patients were categorized into three groups: breast cancer only, viral warts only, and those with both breast cancer and viral warts. The Cox proportion hazard regression analysis was used to measure the effect of HPV on the time to breast cancer diagnosis. Multivariable analyzes and stratified analyzes using hazard ratios (HRs) were presented with 95% confidence intervals (CIs) after adjusting for age, and CCI. RESULT: Among 807,578 HPV population, we identified 6014 breast cancer cases. The HPV group was associated with a significantly higher risk of developing breast cancer (HR, 1.18; 95% CI, 1.15-1.21; p< 0.001) compared with the non-HPV group. HPV patients with age group 18-39 was slightly higher risk of breast cancer occurrence (HR, 1.07; 95% CI, 1.01-1.13; p<.05). The risk of breast cancer in 10-year incidence was 7% higher for females less than 40 years and 23% for over 40 year's patients when compared with non-HPV patients of the same age group. CONCLUSION: Our study indicates that women who develop viral warts are at a significantly higher risk of developing breast cancer than women who have not diagnosed with viral warts. Thus, the presence of viral warts is a potential risk to breast cancer. Therefore, we suggest patients diagnosed with viral warts may get early screening for breast cancer.

A personalized medication management platform (PMMP) to improve medication adherence: A randomized control trial.

OBJECTIVES: Medication non-adherence caused by forgetting and delays has serious health implications and causes substantial expenses to patients, healthcare providers, and insurance companies. We assessed the effectiveness of a personalized medication management platform (PMMP) for improving medication adherence, self-management medication, and reducing long-term medication costs. METHODS: We developed a mobile PMMP to reduce delayed and missed medications. A randomized control trial was conducted of three medical centers in Taiwan. A total 1198 participants who aged over 20 years, received outpatient prescription drugs for a maximum period of 14 days. 763 patients were randomly assigned to intervention group as receiving daily SMS reminders for their medications and 434 patients in control group did not. The primary outcome was change in delaying and forgetting medication between before and after intervention (after 7 days). RESULTS: Medication delays were reduced from 85% to 18% (67% improvement) after SMSs for the intervention group and from 80% to 43% (37% improvement) for the control group. Patients forgot medications were significantly reduced from 46% to 5% (41% improvement) for the experimental group after SMSs and from 44% to 17% (27% improvement) for the control group. The SMSs were considered helpful by 83% of patients and 74% of them thought SMSs help in controlling diseases. 92% of patients would recommend this system to their family and friends. CONCLUSIONS: A timely and personalized medication reminder through SMS can improve medication adherence in a nationalized healthcare system with overall savings in medication costs and significant improvements in health and disease management. TRIAL REGISTRATION: ClinicalTrials.gov: NCT02197689.

Cancer-disease associations: A visualization and animation through medical big data.

OBJECTIVE: Cancer is the primary disease responsible for death and disability worldwide. Currently, prevention and early detection represents the best hope for cure. Knowing the expected diseases that occur with a particular cancer in advance could lead to physicians being able to better tailor their treatment for cancer. The aim of this study was to build an animated visualization tool called as Cancer Associations Map Animation (CAMA), to chart the association of cancers with other disease over time. METHODS: The study population was collected from the Taiwan National Health Insurance Database during the period January 2000 to December 2002, 782 million outpatient visits were used to compute the associations of nine major cancers with other diseases. A motion chart was used to quantify and visualize the associations between diseases and cancers. RESULTS: The CAMA motion chart that was built successfully facilitated the observation of cancer-disease associations across ages and genders. The CAMA system can be accessed online at http://203.71.86.98/web/runq16.html. CONCLUSION: The CAMA animation system is an animated medical data visualization tool which provides a dynamic, time-lapse, animated view of cancer-disease associations across different age groups and gender. Derived from a large, nationwide healthcare dataset, this exploratory data analysis tool can detect cancer comorbidities earlier than is possible by manual inspection. Taking into account the trajectory of cancer-specific comorbidity development may facilitate clinicians and healthcare researchers to more efficiently explore early stage hypotheses, develop new cancer treatment approaches, and identify potential effect modifiers or new risk factors associated with specific cancers.

A richly interactive exploratory data analysis and visualization tool using electronic medical records.

BACKGROUND: Electronic medical records (EMRs) contain vast amounts of data that is of great interest to physicians, clinical researchers, and medial policy makers. As the size, complexity, and accessibility of EMRs grow, the ability to extract meaningful information from them has become an increasingly important problem to solve. METHODS: We develop a standardized data analysis process to support cohort study with a focus on a particular disease. We use an interactive divide-and-conquer approach to classify patients into relatively uniform within each group. It is a repetitive process enabling the user to divide the data into homogeneous subsets that can be visually examined, compared, and refined. The final visualization was driven by the transformed data, and user feedback direct to the corresponding operators which completed the repetitive process. The output results are shown in a Sankey diagram-style timeline, which is a particular kind of flow diagram for showing factors' states and transitions over time. RESULTS: This paper presented a visually rich, interactive web-based application, which could enable researchers to study any cohorts over time by using EMR data. The resulting visualizations help uncover hidden information in the data, compare differences between patient groups, determine critical factors that influence a particular disease, and help direct further analyses. We introduced and demonstrated this tool by using EMRs of 14,567 Chronic Kidney Disease (CKD) patients. CONCLUSIONS: We developed a visual mining system to support exploratory data analysis of multi-dimensional categorical EMR data. By using CKD as a model of disease, it was assembled by automated correlational analysis and human-curated visual evaluation. The visualization methods such as Sankey diagram can reveal useful knowledge about the particular disease cohort and the trajectories of the disease over time.

A novel tool for visualizing chronic kidney disease associated polymorbidity: a 13-year cohort study in Taiwan.

OBJECTIVE: The aim of this study is to analyze and visualize the polymorbidity associated with chronic kidney disease (CKD). The study shows diseases associated with CKD before and after CKD diagnosis in a time-evolutionary type visualization. MATERIALS AND METHODS: Our sample data came from a population of one million individuals randomly selected from the Taiwan National Health Insurance Database, 1998 to 2011. From this group, those patients diagnosed with CKD were included in the analysis. We selected 11 of the most common diseases associated with CKD before its diagnosis and followed them until their death or up to 2011. We used a Sankey-style diagram, which quantifies and visualizes the transition between pre- and post-CKD states with various lines and widths. The line represents groups and the width of a line represents the number of patients transferred from one state to another. RESULTS: The patients were grouped according to their states: that is, diagnoses, hemodialysis/transplantation procedures, and events such as death. A Sankey diagram with basic zooming and planning functions was developed that temporally and qualitatively depicts they had amid change of comorbidities occurred in pre- and post-CKD states. DISCUSSION: This represents a novel visualization approach for temporal patterns of polymorbidities associated with any complex disease and its outcomes. The Sankey diagram is a promising method for visualizing complex diseases and exploring the effect of comorbidities on outcomes in a time-evolution style. CONCLUSIONS: This type of visualization may help clinicians foresee possible outcomes of complex diseases by considering comorbidities that the patients have developed.