Objectives: recently, there has been a government-level movement to guarantee the rights of individual entities regarding the use of their personal data worldwide. This movement has been specifically named as ‘MyData’ in South Korea and has variants such as ‘Self data’, ‘Midata’, ‘MesInfos’, ‘Personal Information Management Services’, ‘Personal Data Economy’ and ‘Internet of Me’ in other countries. This research project aimed to establish and demonstrate a system called ‘HiMD’, which allows individuals to select data sharing institutions and control a range of data utilization parameters under the MyData ecosystem. Method: we developed the MyData Platform, a personal health record data sharing system. The HiMD included several user-empowerment functions such as self-determination for data sharing. Actual platform users were recruited from three university-level hospitals for system assessment. Result: females comprised the majority of users with 991 participants (78.1%). Additionally, data consensus results revealed a decrease in given user permissions (from 94.9% to 79.4%) as the range and depth of permissions increased. Most users agreed to open their medical data for commercial uses (n = 1007, 79.4%) and most of those users were interested in DNA analysis (n = 888, 81.2%). Finally, all results for the five questions presented positive answers. All average values on the five questions scored above three on the Likert scale. In other words, over 50% participants gave point 4 and point 5. Conclusion: the contribution of this study is that it developed and demonstrated a MyData system that reflects the right to data portability. It means that the users can proactively make decisions regarding sharing and transferring their own data. These results are expected to contribute to developing future personal health record (PHR) systems of user-oriented and utilization of personal health data.

Background: Postoperative acute kidney injury (AKI) is a significant risk associated with surgeries under general anesthesia, often leading to increased mortality and morbidity. Existing predictive models for postoperative AKI are usually limited to specific surgical areas or require external validation. Objective: We proposed to build a prediction model for postoperative AKI using several machine learning methods. Methods: We conducted a retrospective cohort analysis of noncardiac surgeries from 2009 to 2019 at seven university hospitals in South Korea. We evaluated six machine learning models: deep neural network, logistic regression, decision tree, random forest, light gradient boosting machine, and naïve Bayes for predicting postoperative AKI, defined as a significant increase in serum creatinine or the initiation of renal replacement therapy within 30 days after surgery. The performance of the models was analyzed using the area under the curve (AUC) of the receiver operating characteristic curve, accuracy, precision, sensitivity (recall), specificity, and F1-score. Results: Among the 239,267 surgeries analyzed, 7935 cases of postoperative AKI were identified. The models, using 38 preoperative predictors, showed that deep neural network (AUC=0.832), light gradient boosting machine (AUC=0.836), and logistic regression (AUC=0.825) demonstrated superior performance in predicting AKI risk. The deep neural network model was then developed into a user-friendly website for clinical use. Conclusions: Our study introduces a robust, high-performance AKI risk prediction system that is applicable in clinical settings using preoperative data. This model’s integration into a user-friendly website enhances its clinical utility, offering a significant step forward in personalized patient care and risk management.

Healthcare data has economic value and is evaluated as such. Therefore, it attracted global attention from observational and clinical studies alike. Recently, the importance of data quality research emerged in healthcare data research. Various studies are being conducted on this topic. In this study, we propose a DQ4HEALTH model that can be applied to healthcare when reviewing existing data quality literature. The model includes 5 dimensions and 415 validation rules. The four evaluation indicators include the net pass rate (NPR), weighted pass rate (WPR), net dimensional pass rate (NDPR), and weighted dimensional pass rate (WDPR). They were used to evaluate the Observational Medical Outcomes Partnership Common Data Model (OMOP CDM) at three medical institutions. These indicators identify differences in data quality between the institutions. The NPRs of the three institutions (A, B, and C) were 96.58%, 90.08%, and 90.87%, respectively, and the WPR was 98.52%, 94.26%, and 94.81%, respectively. In the quality evaluation of the dimensions, the consistency was 70.06% of the total error data. The WDPRs were 98.22%, 94.74%, and 95.05% for institutions A, B, and C, respectively. This study presented indices for comparing quality evaluation models and quality in the healthcare field. Using these indices, medical institutions can evaluate the quality of their data and suggest practical directions for decreasing errors.

Background and Objectives: Intraoperative hypotension (IOH) is of great importance in preventing diseases such as postoperative myocardial infarction, acute kidney injury, and mortality. This study aimed to develop and validate machine learning and deep learning models that predict IOH using both biosignals and personalized clinical information for each patient. Materials and Methods: In this retrospective observational study, we used the VitalDB open dataset, which included intraoperative biosignals and clinical information from 6388 patients who underwent non-cardiac surgery between June 2016 and August 2017 at Seoul National University Hospital, Seoul, South Korea. The predictive performances of models trained with four waveforms (arterial blood pressure, electrocardiography, photoplethysmography, and capnography) and clinical information were evaluated and compared at time points at 5 min before the hypotensive event. To predict hypotensive events during surgery, we developed two predictive models: machine learning and deep learning. In total, 2611 patients were enrolled in this retrospective study. Machine and deep learning algorithms were developed and validated using raw waveforms and clinical information as inputs. Results: Gradient boosting machine showed predicted IOH with an AUROC and accuracy of 0.94 (0.93–0.95) and 0.88 (0.86–0.89). A hybrid CNN-RNN model also showed similar performance with an AUROC and accuracy of 0.94 (0.93–0.95) and 0.88 (0.87–0.89). Conclusions: This study developed and validated machine and deep learning models to predict IOH using waveform data and covariate values. In the future, we anticipate that the results of our study will contribute to predicting IOH in real time in the operating room and reducing the occurrence of IOH.

BACKGROUND Postoperative acute kidney injury (AKI) is a significant risk associated with surgeries under general anesthesia, often leading to increased mortality and morbidity. Existing predictive models for postoperative AKI are usually limited to specific surgical areas or require external validation. OBJECTIVE We proposed to build a prediction model for postoperative AKI using several machine learning methods. METHODS We conducted a retrospective cohort analysis of noncardiac surgeries from 2009 to 2019 at seven university hospitals in South Korea. We evaluated six machine learning models: deep neural network, logistic regression, decision tree, random forest, light gradient boosting machine, and naïve Bayes for predicting postoperative AKI, defined as a significant increase in serum creatinine or the initiation of renal replacement therapy within 30 days after surgery. The performance of the models was analyzed using the area under the curve (AUC) of the receiver operating characteristic curve, accuracy, precision, sensitivity (recall), specificity, and F1-score. RESULTS Among the 239,267 surgeries analyzed, 7935 cases of postoperative AKI were identified. The models, using 38 preoperative predictors, showed that deep neural network (AUC=0.832), light gradient boosting machine (AUC=0.836), and logistic regression (AUC=0.825) demonstrated superior performance in predicting AKI risk. The deep neural network model was then developed into a user-friendly website for clinical use. CONCLUSIONS Our study introduces a robust, high-performance AKI risk prediction system that is applicable in clinical settings using preoperative data. This model’s integration into a user-friendly website enhances its clinical utility, offering a significant step forward in personalized patient care and risk management.