Prediction model for myocardial injury after non-cardiac surgery using machine learning.

Myocardial injury after non-cardiac surgery (MINS) is strongly associated with postoperative outcomes. We developed a prediction model for MINS and have provided it online. Between January 2010 and June 2019, a total of 6811 patients underwent non-cardiac surgery with normal preoperative level of cardiac troponin (cTn). We used machine learning techniques with an extreme gradient boosting algorithm to evaluate the effects of variables on MINS development. We generated two prediction models based on the top 12 and 6 variables. MINS was observed in 1499 (22.0%) patients. The top 12 variables in descending order according to the effects on MINS are preoperative cTn level, intraoperative inotropic drug infusion, operation duration, emergency operation, operation type, age, high-risk surgery, body mass index, chronic kidney disease, coronary artery disease, intraoperative red blood cell transfusion, and current alcoholic use. The prediction models are available at https://sjshin.shinyapps.io/mins_occur_prediction/ . The estimated thresholds were 0.47 in 12-variable models and 0.53 in 6-variable models. The areas under the receiver operating characteristic curves are 0.78 (95% confidence interval [CI] 0.77-0.78) and 0.77 (95% CI 0.77-0.78), respectively, with an accuracy of 0.97 for both models. Using machine learning techniques, we demonstrated prediction models for MINS. These models require further verification in other populations.

Cycle-consistent adversarial networks improves generalizability of radiomics model in grading meningiomas on external validation.

The heterogeneity of MRI is one of the major reasons for decreased performance of a radiomics model on external validation, limiting the model's generalizability and clinical application. We aimed to establish a generalizable radiomics model to predict meningioma grade on external validation through leveraging Cycle-Consistent Adversarial Networks (CycleGAN). In this retrospective study, 257 patients with meningioma were included in the institutional training set. Radiomic features (n = 214) were extracted from T2-weighted (T2) and contrast-enhanced T1 (T1C) images. After radiomics feature selection, extreme gradient boosting classifiers were developed. The models were validated in the external validation set consisting of 61 patients with meningiomas. To reduce the gap in generalization associated with the inter-institutional heterogeneity of MRI, the smaller image set style of the external validation was translated into the larger image set style of the institutional training set using CycleGAN. On external validation before CycleGAN application, the performance of the combined T2 and T1C models showed an area under the curve (AUC), accuracy, and F1 score of 0.77 (95% confidence interval 0.63-0.91), 70.7%, and 0.54, respectively. After applying CycleGAN, the performance of the combined T2 and T1C models increased, with an AUC, accuracy, and F1 score of 0.83 (95% confidence interval 0.70-0.97), 73.2%, and 0.59, respectively. Quantitative metrics (by Fréchet Inception Distance) showed that CycleGAN can decrease inter-institutional image heterogeneity while preserving predictive information. In conclusion, leveraging CycleGAN may be helpful to increase the generalizability of a radiomics model in differentiating meningioma grade on external validation.

Analysis of Dual Combination Therapies Used in Treatment of Hypertension in a Multinational Cohort.

Importance: More than 1 billion adults have hypertension globally, of whom 70% cannot achieve their hypertension control goal with monotherapy alone. Data are lacking on clinical use patterns of dual combination therapies prescribed to patients who escalate from monotherapy. Objective: To investigate the most common dual combinations prescribed for treatment escalation in different countries and how treatment use varies by age, sex, and history of cardiovascular disease. Design, Setting, and Participants: This cohort study used data from 11 electronic health record databases that cover 118 million patients across 8 countries and regions between January 2000 and December 2019. Included participants were adult patients (ages ≥18 years) who newly initiated antihypertensive dual combination therapy after escalating from monotherapy. There were 2 databases included for 3 countries: the Iqvia Longitudinal Patient Database (LPD) Australia and Electronic Practice-based Research Network 2019 linked data set from South Western Sydney Local Health District (ePBRN SWSLHD) from Australia, Ajou University School of Medicine (AUSOM) and Kyung Hee University Hospital (KHMC) databases from South Korea, and Khoo Teck Puat Hospital (KTPH) and National University Hospital (NUH) databases from Singapore. Data were analyzed from June 2020 through August 2021. Exposures: Treatment with dual combinations of the 4 most commonly used antihypertensive drug classes (angiotensin-converting enzyme inhibitor [ACEI] or angiotensin receptor blocker [ARB]; calcium channel blocker [CCB]; ß-blocker; and thiazide or thiazide-like diuretic). Main Outcomes and Measures: The proportion of patients receiving each dual combination regimen, overall and by country and demographic subgroup. Results: Among 970 335 patients with hypertension who newly initiated dual combination therapy included in the final analysis, there were 11 494 patients from Australia (including 9291 patients in Australia LPD and 2203 patients in ePBRN SWSLHD), 6980 patients from South Korea (including 6029 patients in Ajou University and 951 patients in KHMC), 2096 patients from Singapore (including 842 patients in KTPH and 1254 patients in NUH), 7008 patients from China, 8544 patients from Taiwan, 103 994 patients from France, 76 082 patients from Italy, and 754 137 patients from the US. The mean (SD) age ranged from 57.6 (14.8) years in China to 67.7 (15.9) years in the Singapore KTPH database, and the proportion of patients by sex ranged from 24 358 (36.9%) women in Italy to 408 964 (54.3%) women in the US. Among 12 dual combinations of antihypertensive drug classes commonly used, there were significant variations in use across country and patient subgroup. For example starting an ACEI or ARB monotherapy followed by a CCB (ie, ACEI or ARB + CCB) was the most commonly prescribed combination in Australia (698 patients in ePBRN SWSLHD [31.7%] and 3842 patients in Australia LPD [41.4%]) and Singapore (216 patients in KTPH [25.7%] and 439 patients in NUH [35.0%]), while in South Korea, CCB + ACEI or ARB (191 patients in KHMC [20.1%] and 1487 patients in Ajou University [24.7%]), CCB + ß-blocker (814 patients in Ajou University [13.5%] and 217 patients in KHMC [22.8%]), and ACEI or ARB + CCB (147 patients in KHMC [15.5%] and 1216 patients in Ajou University [20.2%]) were the 3 most commonly prescribed combinations. The distribution of 12 dual combination therapies were significantly different by age and sex in almost all databases. For example, use of ACEI or ARB + CCB varied from 873 of 3737 patients ages 18 to 64 years (23.4%) to 343 of 2292 patients ages 65 years or older (15.0%) in South Korea's Ajou University database (P for database distribution by age < .001), while use of ACEI or ARB + CCB varied from 2121 of 4718 (44.8%) men to 1721 of 4549 (37.7%) women in Australian LPD (P for drug combination distributions by sex < .001). Conclusions and Relevance: In this study, large variation in the transition between monotherapy and dual combination therapy for hypertension was observed across countries and by demographic group. These findings suggest that future research may be needed to investigate what dual combinations are associated with best outcomes for which patients.

Renin-Angiotensin-Aldosterone System Inhibitors and Risk of Cancer: A Population-Based Cohort Study Using a Common Data Model.

Studies have reported conflicting results on the association between the use of renin-angiotensin-aldosterone system (RAAS) inhibitors and cancer development. We compared the incidence of cancer between patients using RAAS inhibitors and other antihypertensive drugs. This retrospective observational cohort study used data from seven hospitals in Korea that were converted for use in the Observational Medical Outcomes Partnership Common Data Model. A total of 166,071 patients on antihypertensive therapy across the databases of the seven hospitals were divided into two groups according to the use of RAAS inhibitors. The primary outcome was the occurrence of cancer. A total of 166,071 patients across the databases of the seven hospitals was included in the final analysis; 26,650 (16%) were in the RAAS inhibitors group and 139,421 (84%) in the other antihypertensive drugs group. The meta-analysis of the whole cohort showed a lower incidence of cancer occurrence in the RAAS inhibitor group (9.90 vs. 13.28 per 1000 person years; HR, 0.81; 95% confidence interval [CI], 0.75-0.88). After propensity-score matching, the RAAS inhibitor group consistently showed a lower incidence of cancer (9.90 vs. 13.28 per 1000 person years; HR, 0.86; 95% CI, 0.81-0.91). The patients using RAAS inhibitors showed a lower incidence of cancer compared with those using other antihypertensive drugs. These findings support the association between the use of RAAS inhibitors and cancer occurrence.

Prediction model for postoperative atrial fibrillation in non-cardiac surgery using machine learning.

Some patients with postoperative atrial fibrillation (POAF) after non-cardiac surgery need treatment, and a predictive model for these patients is clinically useful. Here, we developed a predictive model for POAF in non-cardiac surgery based on machine learning techniques. In a total of 201,864 patients who underwent non-cardiac surgery between January 2011 and June 2019 at our institution, 5,725 (2.8%) were treated for POAF. We used machine learning with an extreme gradient boosting algorithm to evaluate the effects of variables on POAF. Using the top five variables from this algorithm, we generated a predictive model for POAF and conducted an external validation. The top five variables selected for the POAF model were age, lung operation, operation duration, history of coronary artery disease, and hypertension. The optimal threshold of probability in this model was estimated to be 0.1, and the area under the receiver operating characteristic (AUROC) curve was 0.80 with a 95% confidence interval of 0.78-0.81. Accuracy of the model using the estimated threshold was 0.95, with sensitivity and specificity values of 0.28 and 0.97, respectively. In an external validation, the AUROC was 0.80 (0.78-0.81). The working predictive model for POAF requiring treatment in non-cardiac surgery based on machine learning techniques is provided online (https://sjshin.shinyapps.io/afib_predictor_0913/). The model needs further verification among other populations.

Predictability of Mortality in Patients With Myocardial Injury After Noncardiac Surgery Based on Perioperative Factors via Machine Learning: Retrospective Study.

BACKGROUND: Myocardial injury after noncardiac surgery (MINS) is associated with increased postoperative mortality, but the relevant perioperative factors that contribute to the mortality of patients with MINS have not been fully evaluated. OBJECTIVE: To establish a comprehensive body of knowledge relating to patients with MINS, we researched the best performing predictive model based on machine learning algorithms. METHODS: Using clinical data from 7629 patients with MINS from the clinical data warehouse, we evaluated 8 machine learning algorithms for accuracy, precision, recall, F1 score, area under the receiver operating characteristic (AUROC) curve, and area under the precision-recall curve to investigate the best model for predicting mortality. Feature importance and Shapley Additive Explanations values were analyzed to explain the role of each clinical factor in patients with MINS. RESULTS: Extreme gradient boosting outperformed the other models. The model showed an AUROC of 0.923 (95% CI 0.916-0.930). The AUROC of the model did not decrease in the test data set (0.894, 95% CI 0.86-0.922; P=.06). Antiplatelet drugs prescription, elevated C-reactive protein level, and beta blocker prescription were associated with reduced 30-day mortality. CONCLUSIONS: Predicting the mortality of patients with MINS was shown to be feasible using machine learning. By analyzing the impact of predictors, markers that should be cautiously monitored by clinicians may be identified.

Style transfer strategy for developing a generalizable deep learning application in digital pathology.

BACKGROUND AND OBJECTIVES: Despite recent advances in artificial intelligence for medical images, the development of a robust deep learning model for identifying malignancy on pathology slides has been limited by problems related to substantial inter- and intra-institutional heterogeneity attributable to tissue preparation. The paucity of available data aggravates this limitation for relatively rare cancers. Here, using ovarian cancer pathology images, we explored the effect of image-to-image style transfer approaches on diagnostic performance. METHODS: We leveraged a relatively large public image set for 142 patients with ovarian cancer from The Cancer Image Archive (TCIA) to fine-tune the renowned deep learning model Inception V3 for identifying malignancy on tissue slides. As an external validation, the performance of the developed classifier was tested using a relatively small institutional pathology image set for 32 patients. To reduce deterioration of the performance associated with the inter-institutional heterogeneity of pathology slides, we translated the style of the small image set of the local institution into the large image set style of the TCIA using cycle-consistent generative adversarial networks. RESULTS: Without style transfer, the performance of the classifier was as follows: area under the receiver operating characteristic curve (AUROC) = 0.737 and area under the precision recall curve (AUPRC) = 0.710. After style transfer, AUROC and AUPRC improved to 0.916 and 0.898, respectively. CONCLUSIONS: This study provides a case of the successful application of style transfer technology to generalize a deep learning model into small image sets in the field of digital pathology. Researchers at local institutions can select this collaborative system to make their small image sets acceptable to the deep learning model.

Network Analysis of Citation in Hypertension Clinical Guidelines.

Recently the two most influential clinical guideline were published for diagnosing and treating hypertension in US and Europe: 2017 American College of Cardiology/American Heart Association (ACC/AHA) and 2018 European Society of Cardiology/European Society of Hypertension (ESC/ESH) Guideline. Though both of them have most in common, the differences in details between guidelines have confused many clinicians in the world. Because guidelines were evidence- based literature, through the analysis of articles cited in guidelines, these similarities and differences could be explained. Bibliometric analysis is a method of quantifying the contents of literature to analyze literature. So using the bibliometric analysis including co-citation network analysis, articles cited in guideline were analyzed. As a result, we figured out that bibliometrics can analyze the influence of the countries, authors and studies on the guidelines, which might affect on the similarities and the differences between both guidelines.

Genomic Common Data Model for Biomedical Data in Clinical Practice.

A common data model for clinical NGS panel data that is used in a distributed research network to achieve large scale to make evidence for improving patient care should be developed. This study developed OMOP-CDM extension for NGS panel data and confirmed the feasibility of the model by finding the differences between a database generated by research-purpose and clinical practice. We believe this data model can be used in distributed research model and will facilitate the usage of the clinical NGS data in patient care.

Genomic Common Data Model for Seamless Interoperation of Biomedical Data in Clinical Practice: Retrospective Study.

BACKGROUND: Clinical sequencing data should be shared in order to achieve the sufficient scale and diversity required to provide strong evidence for improving patient care. A distributed research network allows researchers to share this evidence rather than the patient-level data across centers, thereby avoiding privacy issues. The Observational Medical Outcomes Partnership (OMOP) common data model (CDM) used in distributed research networks has low coverage of sequencing data and does not reflect the latest trends of precision medicine. OBJECTIVE: The aim of this study was to develop and evaluate the feasibility of a genomic CDM (G-CDM), as an extension of the OMOP-CDM, for application of genomic data in clinical practice. METHODS: Existing genomic data models and sequencing reports were reviewed to extend the OMOP-CDM to cover genomic data. The Human Genome Organisation Gene Nomenclature Committee and Human Genome Variation Society nomenclature were adopted to standardize the terminology in the model. Sequencing data of 114 and 1060 patients with lung cancer were obtained from the Ajou University School of Medicine database of Ajou University Hospital and The Cancer Genome Atlas, respectively, which were transformed to a format appropriate for the G-CDM. The data were compared with respect to gene name, variant type, and actionable mutations. RESULTS: The G-CDM was extended into four tables linked to tables of the OMOP-CDM. Upon comparison with The Cancer Genome Atlas data, a clinically actionable mutation, p.Leu858Arg, in the EGFR gene was 6.64 times more frequent in the Ajou University School of Medicine database, while the p.Gly12Xaa mutation in the KRAS gene was 2.02 times more frequent in The Cancer Genome Atlas dataset. The data-exploring tool GeneProfiler was further developed to conduct descriptive analyses automatically using the G-CDM, which provides the proportions of genes, variant types, and actionable mutations. GeneProfiler also allows for querying the specific gene name and Human Genome Variation Society nomenclature to calculate the proportion of patients with a given mutation. CONCLUSIONS: We developed the G-CDM for effective integration of genomic data with standardized clinical data, allowing for data sharing across institutes. The feasibility of the G-CDM was validated by assessing the differences in data characteristics between two different genomic databases through the proposed data-exploring tool GeneProfiler. The G-CDM may facilitate analyses of interoperating clinical and genomic datasets across multiple institutions, minimizing privacy issues and enabling researchers to better understand the characteristics of patients and promote personalized medicine in clinical practice.