Search

BackgroundMachine learning (ML) is now widely deployed in our everyday lives. Building robust ML models requires a massive amount of data for training. Traditional ML algorithms require training data centralization, which raises privacy and data governance issues. Federated learning (FL) is an approach to overcome this issue. We focused on applying FL on vertically partitioned data, in which an individual’s record is scattered among different sites. ObjectiveThe aim of this study was to perform FL on vertically partitioned data to achieve performance comparable to that of centralized models without exposing the raw data. MethodsWe used three different datasets (Adult income, Schwannoma, and eICU datasets) and vertically divided each dataset into different pieces. Following the vertical division of data, overcomplete autoencoder-based model training was performed for each site. Following training, each site’s data were transformed into latent data, which were aggregated for training. A tabular neural network model with categorical embedding was used for training. A centrally based model was used as a baseline model, which was compared to that of FL in terms of accuracy and area under the receiver operating characteristic curve (AUROC). ResultsThe autoencoder-based network successfully transformed the original data into latent representations with no domain knowledge applied. These altered data were different from the original data in terms of the feature space and data distributions, indicating appropriate data security. The loss of performance was minimal when using an overcomplete autoencoder; accuracy loss was 1.2%, 8.89%, and 1.23%, and AUROC loss was 1.1%, 0%, and 1.12% in the Adult income, Schwannoma, and eICU dataset, respectively. ConclusionsWe proposed an autoencoder-based ML model for vertically incomplete data. Since our model is based on unsupervised learning, no domain-specific knowledge is required in individual sites. Under the circumstances where direct data sharing is not available, our approach may be a practical solution enabling both data protection and building a robust model.

BackgroundPrivacy should be protected in medical data that include patient information. A distributed research network (DRN) is one of the challenges in privacy protection and in the encouragement of multi-institutional clinical research. A DRN standardizes multi-institutional data into a common structure and terminology called a common data model (CDM), and it only shares analysis results. It is necessary to measure how a DRN protects patient information privacy even without sharing data in practice. ObjectiveThis study aimed to quantify the privacy risk of a DRN by comparing different deidentification levels focusing on personal health identifiers (PHIs) and quasi-identifiers (QIs). MethodsWe detected PHIs and QIs in an Observational Medical Outcomes Partnership (OMOP) CDM as threatening privacy, based on 18 Health Insurance Portability and Accountability Act of 1996 (HIPPA) identifiers and previous studies. To compare the privacy risk according to the different privacy policies, we generated limited and safe harbor data sets based on 16 PHIs and 12 QIs as threatening privacy from the Synthetic Public Use File 5 Percent (SynPUF5PCT) data set, which is a public data set of the OMOP CDM. With minimum cell size and equivalence class methods, we measured the privacy risk reduction with a trust differential gap obtained by comparing the two data sets. We also measured the gap in randomly sampled records from the two data sets to adjust the number of PHI or QI records. ResultsThe gaps averaged 31.448% and 73.798% for PHIs and QIs, respectively, with a minimum cell size of one, which represents a unique record in a data set. Among PHIs, the national provider identifier had the highest gap of 71.236% (71.244% and 0.007% in the limited and safe harbor data sets, respectively). The maximum size of the equivalence class, which has the largest size of an indistinguishable set of records, averaged 771. In 1000 random samples of PHIs, Device_exposure_start_date had the highest gap of 33.730% (87.705% and 53.975% in the data sets). Among QIs, Death had the highest gap of 99.212% (99.997% and 0.784% in the data sets). In 1000, 10,000, and 100,000 random samples of QIs, Device_treatment had the highest gaps of 12.980% (99.980% and 87.000% in the data sets), 60.118% (99.831% and 39.713%), and 93.597% (98.805% and 5.207%), respectively, and in 1 million random samples, Death had the highest gap of 99.063% (99.998% and 0.934% in the data sets). ConclusionsIn this study, we verified and quantified the privacy risk of PHIs and QIs in the DRN. Although this study used limited PHIs and QIs for verification, the privacy limitations found in this study could be used as a quality measurement index for deidentification of multi-institutional collaboration research, thereby increasing DRN safety.

BackgroundClear guidelines for a patient with suspected COVID-19 infection are unavailable. Many countries rely on assessments through a national hotline or telecommunications, but this only adds to the burden of an already overwhelmed health care system. In this study, we developed an algorithm and a web application to help patients get screened. ObjectiveThis study aims to aid the general public by developing a web-based application that helps patients decide when to seek medical care during a novel disease outbreak. MethodsThe algorithm was developed via consultations with 6 physicians who directly screened, diagnosed, and/or treated patients with COVID-19. The algorithm mainly focused on when to test a patient in order to allocate limited resources more efficiently. The application was designed to be mobile-friendly and deployed on the web. We collected the application usage pattern data from March 1 to March 27, 2020. We evaluated the association between the usage pattern and the numbers of COVID-19 confirmed, screened, and mortality cases by access location and digital literacy by age group. ResultsThe algorithm used epidemiological factors, presence of fever, and other symptoms. In total, 83,460 users accessed the application 105,508 times. Despite the lack of advertisement, almost half of the users accessed the application from outside of Korea. Even though the digital literacy of the 60+ years age group is half of that of individuals in their 50s, the number of users in both groups was similar for our application. ConclusionsWe developed an expert-opinion–based algorithm and web-based application for screening patients. This innovation can be helpful in circumstances where information on a novel disease is insufficient and may facilitate efficient medical resource allocation.

BackgroundAlthough we are living in an era of transparency, medical documents are often still difficult to access. Blockchain technology allows records to be both immutable and transparent. ObjectiveUsing blockchain technology, the aim of this study was to develop a medical document monitoring system that informs patients of changes to their medical documents. We then examined whether patients can effectively verify the monitoring of their primary care clinical medical records in a system based on blockchain technology. MethodsWe enrolled participants who visited two primary care clinics in Korea. Three substudies were performed: (1) a survey of the recognition of blockchain medical records changes and the digital literacy of participants; (2) an observational study on participants using the blockchain-based mobile alert app; and (3) a usability survey study. The participants’ medical documents were profiled with HL7 Fast Healthcare Interoperability Resources, hashed, and transacted to the blockchain. The app checked the changes in the documents by querying the blockchain. ResultsA total of 70 participants were enrolled in this study. Considering their recognition of changes to their medical records, participants tended to not allow these changes. Participants also generally expressed a desire for a medical record monitoring system. Concerning digital literacy, most questions were answered with “good,” indicating fair digital literacy. In the second survey, only 44 participants—those who logged into the app more than once and used the app for more than 28 days—were included in the analysis to determine whether they exhibited usage patterns. The app was accessed a mean of 5.1 (SD 2.6) times for 33.6 (SD 10.0) days. The mean System Usability Scale score was 63.21 (SD 25.06), which indicated satisfactory usability. ConclusionsPatients showed great interest in a blockchain-based system to monitor changes in their medical records. The blockchain system is useful for informing patients of changes in their records via the app without uploading the medical record itself to the network. This ensures the transparency of medical records as well as patient empowerment.

Introduction: End-stage lung disease causes cardiac remodeling and induces electrocardiogram (ECG) changes. On the other way, whether lung transplantation (LTx) in end-stage lung disease patients are associated with ECG change is unknown. The object of this study was to investigate ECG changes before and after LTx in end-stage lung disease patients and whether these changes had clinical significance. Method: This was a single-center retrospective cohort study of 280 end-stage lung disease patients who consecutively underwent LTx at a tertiary referral hospital. ECG findings before LTx and within 1 week and 1, 3, and 6 months after LTx were obtained and analyzed. To find clinical meaning, the ECG at 1 month after LTx was analyzed according to 1-year survival (survivor vs non-survivor groups). Survival data were estimated using the Kaplan–Meier method. Results: Significant differences were observed in the PR interval, QRS duration, QT interval, QTc interval, and heart rate before LTx and 1 month after LTx; the PR interval, QRS duration, QTc interval, and heart rate were decreased. Particularly, the QTc interval was significantly decreased 1 month after LTx, whereas there was no significant change in the QTc interval from 1 to 6 months thereafter. The PR interval, QT interval, QTc interval, and heart rate were significantly different between the survivor and non-survivor groups. The serial changes in QTc interval before LTx and 1 and 3 months after LTx were also significantly different between the survivor and non-survivor groups (p = 0.040 after adjusting for age and body mass index). Upon dividing the patients based on the range of QTc interval change ≤ -8 ms, >-8–10 ms, >10–35 ms, >35 ms), the survival rate was significantly lower in the group whose QTc interval at 1 month after LTx decreased by > 35 m (p = 0.019). Conclusion: LTx in patients with end-stage lung disease may induce ECG changes. Patients whose QTc interval at 1 month after LTx decreased by > 35 ms have a significantly higher 1-year mortality rate. Hence, these ECG changes may have clinical and prognostic significance. [ABSTRACT FROM AUTHOR]

Purpose: This study investigates the impact of varying degrees of immunosuppression on the clinical outcomes of immunocompromised individuals, particularly those with autoimmune diseases or post-solid organ transplant statuses, in the context of COVID-19. By focusing on these highly vulnerable populations, the study underscores the significant health inequalities faced by immunocompromised patients, who experience disproportionately worse outcomes in comparison to the general population. Methods: A retrospective cohort analysis of the K-COV-N dataset was conducted, comparing the effects of immunosuppression in autoimmune and transplant groups with matched control groups. Propensity score matching was employed to minimize inequalities in baseline characteristics, ensuring a more equitable comparison between immunocompromised and non-immunocompromised individuals. Outcomes included COVID-19-related in-hospital mortality, 28-day mortality, ICU admissions, and the need for respiratory support among 323,890 adults in the Republic of Korea. Patients with cancer or other immunosuppressive conditions, such as HIV, were excluded. Subgroup analyses assessed the influence of specific immunosuppressive medications and vaccination extent. Results: Significantly elevated in-hospital mortality was found for patients with autoimmune diseases (adjusted Odds Ratio [aOR] 2.749) and transplant recipients (aOR 7.567), with similar patterns in other outcomes. High-dose steroid use and a greater number of immunosuppressant medications markedly increased the risk of poor outcomes. Vaccination emerged as a protective factor, with a single dose substantially improving outcomes for autoimmune patients and at least two doses necessary for transplant recipients. Conclusions: Immunocompromised patients, particularly those with autoimmune diseases and transplant recipients, are highly vulnerable to severe COVID-19 outcomes. High-dose steroid use and multiple immunosuppressants further increase risks. Vaccination significantly improves outcomes, with at least one dose benefiting autoimmune patients and two doses necessary for transplant recipients. Personalized vaccination schedules based on immunosuppression levels are essential to mitigate healthcare inequalities and improve outcomes, particularly in underserved populations, informing both clinical and public health strategies. [ABSTRACT FROM AUTHOR]

Background: Privacy is of increasing interest in the present big data era, particularly the privacy of medical data. Specifically, differential privacy has emerged as the standard method for preservation of privacy during data analysis and publishing., Objective: Using machine learning techniques, we applied differential privacy to medical data with diverse parameters and checked the feasibility of our algorithms with synthetic data as well as the balance between data privacy and utility., Methods: All data were normalized to a range between -1 and 1, and the bounded Laplacian method was applied to prevent the generation of out-of-bound values after applying the differential privacy algorithm. To preserve the cardinality of the categorical variables, we performed postprocessing via discretization. The algorithm was evaluated using both synthetic and real-world data (from the eICU Collaborative Research Database). We evaluated the difference between the original data and the perturbated data using misclassification rates and the mean squared error for categorical data and continuous data, respectively. Further, we compared the performance of classification models that predict in-hospital mortality using real-world data., Results: The misclassification rate of categorical variables ranged between 0.49 and 0.85 when the value of ε was 0.1, and it converged to 0 as ε increased. When ε was between 10 2 and 10 3 , the misclassification rate rapidly dropped to 0. Similarly, the mean squared error of the continuous variables decreased as ε increased. The performance of the model developed from perturbed data converged to that of the model developed from original data as ε increased. In particular, the accuracy of a random forest model developed from the original data was 0.801, and this value ranged from 0.757 to 0.81 when ε was 10 -1 and 10 4 , respectively., Conclusions: We applied local differential privacy to medical domain data, which are diverse and high dimensional. Higher noise may offer enhanced privacy, but it simultaneously hinders utility. We should choose an appropriate degree of noise for data perturbation to balance privacy and utility depending on specific situations., (©MinDong Sung, Dongchul Cha, Yu Rang Park. Originally published in JMIR Medical Informatics (https://medinform.jmir.org), 08.11.2021.)

According to recent revisions to medical laws in Korea, changes to electronic medical records are to be documented. To do so, however, a transparent system with which to store original documents and changes thereto is needed. The transparency and immutability of blockchain records are the key characters of blockchain technology. Employing these characteristics, we developed an application with which to monitor changes of medical records using blockchain.

Background: We compared the chest configurations of patients with primary spontaneous pneumothorax (PSP) and age-sex-matched controls to determine the presence of chest wall deformities in patients with PSP., Methods: We retrospectively enrolled 166 male patients with PSP (age, 18-19 years) and 85 age-sex-matched controls without PSP, who simultaneously underwent chest computed tomography (CT) and radiography at one of two institutes. After correcting for height, the following thoracic parameters were comparatively evaluated between the two groups: maximal internal transverse ( T ) and anteroposterior ( W ) diameters of the chest, maximal internal lung height ( H ), Haller index ( T / W ), and T /Height, T / H , W /Height, W / H , and H /Height ratios., Results: Patients were taller than the control subjects (176.5 cm ± 5.9 cm versus 174.4 cm ± 5.6 cm; p  = 0.007). After controlling for height, the patient group exhibited lower T and W and greater H and Haller index values than the control group ( T : 95% confidence interval [CI], 24.8-25.2 cm versus 25.9-26.5; W : 95% CI, 8.9-9.2 cm versus 10.1-10.6 cm; H: 95% CI, 25.2-25.9 cm versus 23.4-24.4 cm; and Haller index, 95% CI, 2.7-2.9 versus 2.4-2.6; all, p  < 0.001). The patient group also exhibited lower T /Height, T /H, W /Height, and W /H ratios and greater H /Height ratio than the control group., Conclusions: Patients with PSP have an anteroposteriorly flatter, laterally narrower, and craniocaudally taller thorax than subjects without PSP, suggesting that chest configuration is associated with the development of pneumothorax., Competing Interests: None., (Georg Thieme Verlag KG Stuttgart · New York.)

Background: Active tuberculosis (TB) risk increases during anti-tumor necrosis factor (TNF) therapy; therefore, latent TB infection (LTBI) screening is recommended in potential TNF inhibitor users. It is unclear whether anti-TNF therapy increases the risk of active TB infection even after standard LTBI treatment., Objective: The objective of this study was to compare the risk of active TB development in LTBI-positive versus LBTI-negative TNF inhibitor users following the current national LTBI treatment guidelines for LTBI., Methods: We retrospectively studied 949 TNF inhibitor users with immune-mediated inflammatory diseases from 2005 to 2012 at the Yonsei University Health System. We compared the incidence of active TB among LTBI-positive TNF inhibitor users treated according to national guidelines (n = 256) and LTBI-negative TNF inhibitor users (n = 521), using Poisson regression., Results: The active TB incidence was 1107 per 100,000 patient-years in LTBI-positive TNF inhibitor users who received standard LTBI treatment and 490 per 100,000 patient-years in LTBI-negative TNF inhibitor users. Analysis showed that despite this numerical trend active TB risk was not statistically significantly elevated in LTBI-positive versus LTBI-negative TNF inhibitor users (incidence risk ratio, 2.15; P = 0.24; 95% confidence interval, 0.6-7.7)., Conclusions: This study demonstrated no statistically significantly increased risk of active TB in LTBI-positive TNF inhibitor users who received standard LTBI treatment compared with LTBI-negative TNF inhibitor users.