Search

Artificial intelligence is increasingly being adopted in medical fields to predict various outcomes. In particular, chronic kidney disease (CKD) is problematic because it often progresses to end-stage kidney disease. However, the trajectories of kidney function depend on individual patients. In this study, we propose a machine learning-based model to predict the rapid decline in kidney function among CKD patients by using a big hospital database constructed from the information of 118,584 patients derived from the electronic medical records system. The database included the estimated glomerular filtration rate (eGFR) of each patient, recorded at least twice over a period of 90 days. The data of 19,894 patients (16.8%) were observed to satisfy the CKD criteria. We characterized the rapid decline of kidney function by a decline of 30% or more in the eGFR within a period of two years and classified the available patients into two groups—those exhibiting rapid eGFR decline and those exhibiting non-rapid eGFR decline. Following this, we constructed predictive models based on two machine learning algorithms. Longitudinal laboratory data including urine protein, blood pressure, and hemoglobin were used as covariates. We used longitudinal statistics with a baseline corresponding to 90-, 180-, and 360-day windows prior to the baseline point. The longitudinal statistics included the exponentially smoothed average (ESA), where the weight was defined to be 0.9*(t/b), where t denotes the number of days prior to the baseline point and b denotes the decay parameter. In this study, b was taken to be 7 (7-day ESA). We used logistic regression (LR) and random forest (RF) algorithms based on Python code with scikit-learn library (https://scikit-learn.org/) for model creation. The areas under the curve for LR and RF were 0.71 and 0.73, respectively. The 7-day ESA of urine protein ranked within the first two places in terms of importance according to both models. Further, other features related to urine protein were likely to rank higher than the rest. The LR and RF models revealed that the degree of urine protein, especially if it exhibited an increasing tendency, served as a prominent risk factor associated with rapid eGFR decline. [ABSTRACT FROM AUTHOR]

Access control policies for XML typically use regular path expressions such as XPath for specifying the objects for access-control policies. However such access-control policies are burdens to the query engines for XML documents. To relieve this burden, we introduce static analysis for XML access-control. Given an access-control policy, query expression, and an optional schema, static analysis determines if this query expression is guaranteed not to access elements or attributes that are hidden by the access-control policy but permitted by the schema. Static analysis can be performed without evaluating any query expression against actual XML documents. Run-time checking is required only when static analysis is unable to determine whether to grant or deny access requests. A side effect of static analysis is query optimization: access-denied expressions in queries can be evaluated to empty lists at compile time. We further extend static analysis for handling value-based access-control policies and introduce view schemas. [ABSTRACT FROM AUTHOR]

Over the years a wide variety of access control models and policies have been proposed, and almost all the models have assumed "grant the access request or deny it." They do not provide any mechanism that enables us to bind authorization rules with required operations such as logging and encryption. We propose the notion of a "provisional action" that tells the user that his request will be authorized provided he (and/or the system) takes certain actions. The major advantage of our approach is that arbitrary actions such as cryptographic operations can all coexist in the access control policy rules. We define a fundamental authorization mechanism and then formalize a provision-based access control model. We also present algorithms and describe their algorithmic complexity. Finally, we illustrate how provisional access control policy rules can be specified effectively in practical usage scenarios. [ABSTRACT FROM AUTHOR]

Background: EMRs have enabled the application of machine learning (ML) to characterize disease progression. ML requires large data to produce meaningful results, thus penalizing health systems with small EMRs. Applying models pre-trained on different EMRs is useful but may also suffer from discrepancies between medical sites using different disease codes, procedures, etc. We study the models trained on a large EMR and applied on a smaller one to estimate T2DM complication time using survival models. Method: We trained survival models on a large US insurance claim dataset to create a right-censored cohort (C_1) that includes onset time of T2DM complications, patient profiles, and disease histories. We constructed features using age, sex, and past occurrences of 250 most frequent Clinical Classification Software codes. We trained Random Survival Forest (RSF) models on this cohort. We then created another cohort (C_2) from a smaller EMR from a large Japanese hospital, on which we trained RSF models and also applied models trained on C_1. Results: Table 1 shows a feature importance ranking of trained RSF models and C-indices results. When applying C_1 pre-trained models to C_2, C-indices are high in Nephrology, Hyperosmolar, and Neuropathy. In these cases, common features are highly ranked by both models. Conclusions: This study illustrates the applicability of survival models with disease code covariates on cohorts from different geos. Disclosure: A. Koseki: Employee; Self; IBM, Research Support; Self; Astellas Pharma Inc. M. Makino: None. A. Suzuki: Research Support; Self; Chugai Pharmaceutical Co., Ltd., Ono Pharmaceutical Co., Ltd., Taisho Pharmaceutical Co., Ltd., Takeda Pharmaceutical Co., Speaker's Bureau; Self; Eli Lilly Japan K. K. R. Tokumasu: None. P. Chakraborty: None. M. Ghalwash: None. T. Iwamori: None. M. Kudo: Employee; Self; IBM. D. Sow: None. H. Yanagisawa: None. R. Yanagiya: None. [ABSTRACT FROM AUTHOR]

Chronic Kidney Disease (CKD) for type 2 diabetes mellitus (T2DM) is an increasingly serious issue worldwide. CKD stage is known to progress in stages where 1 is normal and 5 is the worst case of hemodialysis. We have previously reported on the accuracy of AI technology to target CKD stage 1 patients and predict the progression to stages 2 and above. Such prediction is targeted to help identify patients who require intensive care. Such study showed maximum 0.75 AUC score. In this study, we have extended our study to investigate all other stage combinations. For example, we can set our goal to answer questions such as: "Can AI technology accurately predict stage 2 patients progressing into stage 5 (hemodialysis)?" We have investigated the EMR (Electronic Medical Record) of 64,059 type-2 diabetes patients who have visited the hospital. We have extracted 36 features, where 12 sources (e.g., Urine protein, albuminuria and eGFR) were selectively chosen by extraction from known literature and 3 types of values (mean, latest, SD) were calculated for each source. Period of 180 days range was scanned within 20% range (i.e., 180 days * 0.2 = 36 days). We balanced the label transition of each stage evenly by the number of data for each good label (no progress from current target stage) and bad label (progress from target to worse stage). We created a new prediction model for each combination (i.e., 1 to 2, 1 to 3, 1 to 4, 1 to 5, etc.) and performed cross validation by Logistic Regression to calculate the accuracy, F-Score and AUC. All the combinations showed relatively good results. The range and average of each measurements were: Accuracy 0.653 through 0.753 Average: 0.708. F-Score 0.673 through 0.754 Average:0.719. AUC 0.705 through 0.826 Average 0.770. We applied AI techniques to create new predictive model which can detect the progression for all combinations of type-2 diabetes CKD stage progression. This model may contribute by a more effective and accurate intervention to reduce hemodialysis and cardiovascular events. Disclosure: K. Miyamoto: None. A. Koseki: Employee; Self; IBM. M. Kudo: Employee; Self; IBM. M. Makino: None. A. Suzuki: Research Support; Self; Chugai Pharmaceutical Co., Ltd., Kowa Company, Ltd., Ono Pharmaceutical Co., Ltd., Taisho Pharmaceutical Co., Ltd., Takeda Pharmaceutical Company Limited. [ABSTRACT FROM AUTHOR]

Background: In the presence of a large data set of electronic health records (EHRs), predicting the future diabetic complications is of importance for decision making in the medical treatments. Using modern machine learning techniques, it is generally becoming easier to build complex models to predict the future. While complex models are giving good prediction performance, simple models should be more useful in terms of interpretability and practicality in the real medical fields. For example the less the explanatory variables such as lab tests are used for the model, the more useful it is. Our interests thus lie in whether accurate prediction is possible when using less and major explanatory variables to make a model to predict the future diabetic complications. Method: In this paper, we make a prediction model of heart failures for diabetics in half, 1, 3, 5 years using 13 longitudinal lab tests records including hemoglobin A1c, fasting blood glucose, low-density cholesterol, high-density cholesterol, triglycerides, uric acid, serum creatinine, estimated glomerular filtration rate, proteinuria, albuminuria, gamma-glutamyl transpeptidase, alanine transaminase, aspartate transaminase, from which we extracted several longitudinal statistics for input variables. We also compared the prediction performance using several major machine learning algorithms. Results: In prediction of future heart failures in half, 1, 3, 5 years, AUC of 0.77, 0.79, 0.79, and 0.80 were marked when using Logistic Regression. Comparing algorithms for half year prediction, ensemble algorithms outperformed Logistic Regression. The AdaBoost algorithm marked AUC of 0.82. Conclusion: We observed that when using only 13 longitudinal lab tests records, future heart failure prediction is successfully made by using modern machine learning algorithms. Disclosure: A. Koseki: Employee; Self; IBM. T. Ohko: Employee; Self; IBM. M. Kudo: Employee; Self; IBM. M. Makino: None. A. Suzuki: Research Support; Self; Chugai Pharmaceutical Co., Ltd., Kowa Company, Ltd., Ono Pharmaceutical Co., Ltd., Taisho Pharmaceutical Co., Ltd., Takeda Pharmaceutical Company Limited. [ABSTRACT FROM AUTHOR]

Artificial intelligence (AI) is expected to support clinical judgement in medicine. We constructed a new predictive model for diabetic kidney diseases (DKD) using AI, processing natural language and longitudinal data with big data machine learning, based on the electronic medical records (EMR) of 64,059 diabetes patients. AI extracted raw features from the previous 6 months as the reference period and selected 24 factors to find time series patterns relating to 6-month DKD aggravation, using a convolutional autoencoder. AI constructed the predictive model with 3,073 features, including time series data using logistic regression analysis. AI could predict DKD aggravation with 71% accuracy. Furthermore, the group with DKD aggravation had a significantly higher incidence of hemodialysis than the non-aggravation group, over 10 years (N = 2,900). The new predictive model by AI could detect progression of DKD and may contribute to more effective and accurate intervention to reduce hemodialysis. [ABSTRACT FROM AUTHOR]

Background: Thanks to technological advancements for medical devices, we can measure glucose by the minute for weeks using a sensor called the continuous glucose monitoring (CGM) system. CGM is time-series data and has been available since devices with low measurement error appeared 10 years ago. CGM can be relied upon to help make treatment decisions. A major issue regarding CGM is clinical interpretation by physicians. Methods: CGM data were obtained by flush glucose monitoring system in 156 type 2 diabetes patients. We divided the patients into 2 groups by mean HbA1c levels during 6 months before getting CGM data. High HbA1c group (High) had their mean HbA1c levels with equal to or above 7% and below 9%. Low HbA1c group (Low) had their mean HbA1c levels with equal to or above 5% and below 7%. The patients with their HbA1c levels equal to or above 9% was excluded from this study. We conducted the experiment with manually created dataset that is designed for evaluating the performance of trajectory extraction. Artificial intelligence (AI) performed pattern mining of raw data of flush glucose in 4 to 8 sequential data sets. Results: Among 156 patients, there were 83 High group patients, while 58 patients were defined as Low group. We excluded 15 patients from this study due to high HbA1c levels, while there was no patient with HbA1c level below 5%. AI constructed 1292 patterns of glucose trajectories from CGM data. We found that 67 patterns were significantly different between High and Low groups 8p<0.05). Conclusion: In this session, we propose a method of extracting the trajectories of glucose values of CGM in type 2 diabetes patients. Our method could contribute to better CGM interpretation. Disclosure: M. Makino: None. R. Yoshimoto: None. M. Kondo-Ando: None. Y. Yoshino: None. I. Hiratsuka: None. W. Maki: None. S. Sekiguchi-Ueda: None. A. Kakita: None. M. Shibata: None. Y. Seino: None. T. Takayanagi: None. M. Ono: None. A. Koseki: Employee; Self; IBM. M. Kudo: Employee; Self; IBM. K. Haida: None. R. Yanagiya: None. N. Hayakawa: None. A. Suzuki: Research Support; Self; Chugai Pharmaceutical Co., Ltd., Dai-ichi Life Insurance Company, IBM, MSD, Ono Pharmaceutical Co., Ltd., Takeda Pharmaceutical Company Limited. Speaker's Bureau; Self; Asahi Kasei Corporation, Daiichi Sankyo Company, Limited, Eli Lilly and Company, Mitsubishi Tanabe Pharma Corporation, Taisho Pharmaceutical Co., Ltd. [ABSTRACT FROM AUTHOR]

Background: In the presence of large data set of electronic health records (EHRs), predicting the future disease status is of importance for decision making in the medical treatments. Using modern machine learning techniques, it is generally becoming easier to build complex models to predict the future. For those models, a set of past information are used to make explanatory variables, however, we don't have enough knowledge as to how long we should collect data backward. In some cases, very late tendencies are influencing the future status of disease while in the other cases, old events were the importance causes of the change of the disease status. Our interest thus lies in how old data we have to process to make the good prediction models. Method: In this paper, we discuss a set of machine learning algorithms to predict the diabetic nephropathy stage in the future using sets of input variables which were collected from different time span of past records. To compare the performance of algorithms we used Logistic Regression, AdaBoost, Gradient Boosting, Decision tree, Multi-layer Perceptron, and Random Forest. We then provide different set of variables of EHR that include past 30-, 60-, 90-, 180-, 210-, 240-, 270-, 300-, 330-, and 360-day data sets, from which we extracted several longitudinal statistics for input variables. From about 65 thousand type 2 diabetes patients, the models classify whether the nephropathy stage gets aggravated or stay in 180 days. Results: For almost all algorithms, AUC is getting improved when using older data, and 360-day data sets gave the best. Among the algorithms, Gradient Boosting gave the best AUC of 0.77 when using 360-day data set. When using 360-day data sets, Decision Tree gave worst AUC of 0.61. Conclusion: We observed that when using to past data up to 360 days, the oldest data set gave the best prediction performance. Longitudinal statistics in rather long span gives good explanatory information for future nephropathy development. Disclosure: A. Koseki: Employee; Self; IBM. M. Ono: None. M. Kudo: Employee; Self; IBM. K. Haida: None. M. Makino: None. A. Suzuki: Research Support; Self; Chugai Pharmaceutical Co., Ltd., Dai-ichi Life Insurance Company, IBM, MSD, Ono Pharmaceutical Co., Ltd., Takeda Pharmaceutical Company Limited. Speaker's Bureau; Self; Asahi Kasei Corporation, Daiichi Sankyo Company, Limited, Eli Lilly and Company, Mitsubishi Tanabe Pharma Corporation, Taisho Pharmaceutical Co., Ltd. [ABSTRACT FROM AUTHOR]

In Chronic Kidney Disease (CKD), kidneys are damaged and lose their ability to filter blood, leading to a plethora of health consequences that end up in dialysis. Despite its prevalence, CKD goes often undetected at early stages. In order to better understand disease progression, we stratified patients with CKD by considering the time to dialysis from diagnosis of early CKD (stages 1 or 2). To achieve this, we first reduced the number of clinical features in a predictive time-to-dialysis model and identified the top important features on a cohort of ∼ 40, 000 CKD patients. The extracted features were used to stratify a subpopulation of 3 , 522 patients that showed anemia and were prescribed for cardiovascular-related drugs and progressed faster to dialysis. On the other side, clustering patients using conventional clustering methods based on their clinical features did not allow such clear interpretation to identify the main factors for leading fast progression to dialysis. To our knowledge this is the first study extracting interpretable features for stratifying a cohort of early CKD patients using time-to-event analysis which could help prevention and the development of new treatments., (©2023 AMIA - All rights reserved.)

Tokyo Olympic and Paralympic Games, postponed for the COVID-19 pandemic, were finally held in the summer of 2021. Just before the games, the Alpha variant was being replaced with the more contagious Delta variant. AY.4 substrain AY.29, which harbors two additional characteristic mutations of 5239C > T (NSP3 Y840Y) and 5514T > C (NSP3 V932A), emerged in Japan and became dominant in Tokyo by the time of the Olympic Games. Variants of SARS-CoV-2 genomes were performed to extract AY.29 Delta substrain samples with 5239C > T and 5514T > C. Phylogenetic analysis was performed to illustrate how AY.29 strains evolved and were introduced into countries abroad. Simultaneously, ancestral searches were performed for the overseas AY.29 samples to identify their origins in Japan using the maximum variant approach. As of January 10, 2022, 118 samples were identified in 20 countries. Phylogenetic analysis and ancestral searches identified 55 distinct introductions into those countries. The United States had 50 samples with 10 distinct introductions, and the United Kingdom had 13 distinct strains introduced in 18 samples. Other countries or regions with multiple introductions were Canada, Germany, South Korea, Hong Kong, Thailand, and the Philippines. Among the 20 countries, most European and North American countries have vaccination rates over 50% and sufficient genomic surveillances are conducted; transmissions seem contained. However, propagation to unvaccinated regions might have caused unfathomable damages. Since samples in those unvaccinated countries are also undersampled with a longer lead time for data sharing, it will take longer to grasp the whole picture. More rigorous departure screenings for the participants from the unvaccinated countries might have been necessary., Competing Interests: TK, RT, and MK were employees of IBM. The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest., (Copyright © 2022 Koyama, Tokumasu, Katayama, Saito, Kudo and Imoto.)

Objectives: Trajectories of estimated glomerular filtration rate (eGFR) decline vary highly among patients with chronic kidney disease (CKD). It is clinically important to identify patients who have high risk for eGFR decline. We aimed to identify clusters of patients with extremely rapid eGFR decline and develop a prediction model using a machine learning approach., Design: Retrospective single-centre cohort study., Settings: Tertiary referral university hospital in Toyoake city, Japan., Participants: A total of 5657 patients with CKD with baseline eGFR of 30 mL/min/1.73 m 2 and eGFR decline of ≥30% within 2 years., Primary Outcome: Our main outcome was extremely rapid eGFR decline. To study-complicated eGFR behaviours, we first applied a variation of group-based trajectory model, which can find trajectory clusters according to the slope of eGFR decline. Our model identified high-level trajectory groups according to baseline eGFR values and simultaneous trajectory clusters. For each group, we developed prediction models that classified the steepest eGFR decline, defined as extremely rapid eGFR decline compared with others in the same group, where we used the random forest algorithm with clinical parameters., Results: Our clustering model first identified three high-level groups according to the baseline eGFR (G1, high GFR, 99.7±19.0; G2, intermediate GFR, 62.9±10.3 and G3, low GFR, 43.7±7.8); our model simultaneously found three eGFR trajectory clusters for each group, resulting in nine clusters with different slopes of eGFR decline. The areas under the curve for classifying the extremely rapid eGFR declines in the G1, G2 and G3 groups were 0.69 (95% CI, 0.63 to 0.76), 0.71 (95% CI 0.69 to 0.74) and 0.79 (95% CI 0.75 to 0.83), respectively. The random forest model identified haemoglobin, albumin and C reactive protein as important characteristics., Conclusions: The random forest model could be useful in identifying patients with extremely rapid eGFR decline., Trial Registration: UMIN 000037476; This study was registered with the UMIN Clinical Trials Registry., Competing Interests: Competing interests: Yes, there are competing interests for one or more authors and I have provided a Competing Interests statement in my manuscript and in the box below., (© Author(s) (or their employer(s)) 2022. Re-use permitted under CC BY-NC. No commercial re-use. See rights and permissions. Published by BMJ.)

Background: Emergence of vaccine-escaping SARS-CoV-2 variants is a serious problem for global public health. The currently rampant Omicron has been shown to possess remarkable vaccine escape; however, the selection pressure exerted by vaccines might pave the way for other escape mutants in the near future. Materials & methods: For detection of neutralizing antibodies, the authors used the recently developed HiBiT-based virus-like particle neutralization test system. Sera after vaccination (two doses of Pfizer/BioNTech mRNA vaccine) were used to evaluate the neutralizing activity against various strains of SARS-CoV-2. Results: Beta+R346K, which was identified in the Philippines in August 2021, exhibited the highest vaccine resistance among the tested mutants. Surprisingly, Mu+K417N mutant exhibited almost no decrease in neutralization. Imdevimab retained efficacy against these strains. Conclusions: Mutations outside the receptor-binding domain contributed to vaccine escape. Both genomic surveillance and phenotypic analysis synergistically accelerate identifications of vaccine-escaping strains.

The successive emergence of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) variants has presented a major challenge in the management of the coronavirus disease (COVID-19) pandemic. There are growing concerns regarding the emerging variants escaping vaccines or therapeutic neutralizing antibodies. In this study, we conducted an epidemiological survey to identify SARS-CoV-2 variants that are sporadically proliferating in vaccine-advanced countries. Subsequently, we created HiBiT-tagged virus-like particles displaying spike proteins derived from the variants to analyze the neutralizing efficacy of the BNT162b2 mRNA vaccine and several therapeutic antibodies. We found that the Mu variant and a derivative of the Delta strain with E484K and N501Y mutations significantly evaded vaccine-elicited neutralizing antibodies. This trend was also observed in the Beta and Gamma variants, although they are currently not prevalent. Although 95.2% of the vaccinees exhibited prominent neutralizing activity against the prototype strain, only 73.8 and 78.6% of the vaccinees exhibited neutralizing activity against the Mu and the Delta derivative variants, respectively. A long-term analysis showed that 88.8% of the vaccinees initially exhibited strong neutralizing activity against the currently circulating Delta strain; the number decreased to 31.6% for the individuals at 6 months after vaccination. Notably, these variants were shown to be resistant to several therapeutic antibodies. Our findings demonstrate the differential neutralization efficacy of the COVID-19 vaccine and monoclonal antibodies against circulating variants, suggesting the need for pandemic alerts and booster vaccinations against the currently prevalent variants., Competing Interests: YY is a current employee of Kanto Chemical Co., Inc. TK, RT, and MK are current employees of IBM. The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest., (Copyright © 2022 Miyakawa, Jeremiah, Yamaoka, Koyama, Tokumasu, Kudo, Kato and Ryo.)

This paper describes a technology for predicting the aggravation of diabetic nephropathy from electronic health record (EHR). For the prediction, we used features extracted from event sequence of lab tests in EHR with a stacked convolutional autoencoder which can extract both local and global temporal information. The extracted features can be interpreted as similarities to a small number of typical sequences of lab tests, that may help us to understand the disease courses and to provide detailed health guidance. In our experiments on real-world EHRs, we confirmed that our approach performed better than baseline methods and that the extracted features were promising for understanding the disease.