scGGAN: single-cell RNA-seq imputation by graph-based generative adversarial network.

Single-cell RNA sequencing (scRNA-seq) data are typically with a large number of missing values, which often results in the loss of critical gene signaling information and seriously limit the downstream analysis. Deep learning-based imputation methods often can better handle scRNA-seq data than shallow ones, but most of them do not consider the inherent relations between genes, and the expression of a gene is often regulated by other genes. Therefore, it is essential to impute scRNA-seq data by considering the regional gene-to-gene relations. We propose a novel model (named scGGAN) to impute scRNA-seq data that learns the gene-to-gene relations by Graph Convolutional Networks (GCN) and global scRNA-seq data distribution by Generative Adversarial Networks (GAN). scGGAN first leverages single-cell and bulk genomics data to explore inherent relations between genes and builds a more compact gene relation network to jointly capture the homogeneous and heterogeneous information. Then, it constructs a GCN-based GAN model to integrate the scRNA-seq, gene sequencing data and gene relation network for generating scRNA-seq data, and trains the model through adversarial learning. Finally, it utilizes data generated by the trained GCN-based GAN model to impute scRNA-seq data. Experiments on simulated and real scRNA-seq datasets show that scGGAN can effectively identify dropout events, recover the biologically meaningful expressions, determine subcellular states and types, improve the differential expression analysis and temporal dynamics analysis. Ablation experiments confirm that both the gene relation network and gene sequence data help the imputation of scRNA-seq data.

Re-ordered fuzzy conformance checking for uncertain clinical records.

Modern hospitals implement clinical pathways to standardize patients' treatments. Conformance checking techniques provide an automated tool to assess whether the actual executions of clinical processes comply with the corresponding clinical pathways. However, clinical processes are typically characterized by a high degree of uncertainty, both in their execution and recording. This paper focuses on uncertainty related to logging clinical processes. The logging of the activities executed during a clinical process in the hospital information system is often performed manually by the involved actors (e.g., the nurses). However, such logging can occur at a different time than the actual execution time, which hampers the reliability of the diagnostics provided by conformance checking techniques. To address this issue, we propose a novel conformance checking algorithm that leverages principles of fuzzy set theory to incorporate experts' knowledge when generating conformance diagnostics. We exploit this knowledge to define a fuzzy tolerance in a time window, which is then used to assess the magnitude of timestamp violations of the recorded activities when evaluating the overall process execution compliance. Experiments conducted on a real-life case study in a Dutch hospital show that the proposed method obtains more accurate diagnostics than the state-of-the-art approaches. We also consider how our diagnostics can be used to stimulate discussion with domain experts on possible strategies to mitigate logging uncertainty in the clinical practice.

Chat-ePRO: Development and pilot study of an electronic patient-reported outcomes system based on ChatGPT.

OBJECTIVE: Chatbots have the potential to improve user compliance in electronic Patient-Reported Outcome (ePRO) system. Compared to rule-based chatbots, Large Language Model (LLM) offers advantages such as simplifying the development process and increasing conversational flexibility. However, there is currently a lack of practical applications of LLMs in ePRO systems. Therefore, this study utilized ChatGPT to develop the Chat-ePRO system and designed a pilot study to explore the feasibility of building an ePRO system based on LLM. MATERIALS AND METHODS: This study employed prompt engineering and offline knowledge distillation to design a dialogue algorithm and built the Chat-ePRO system on the WeChat Mini Program platform. In order to compare Chat-ePRO with the form-based ePRO and rule-based chatbot ePRO used in previous studies, we conducted a pilot study applying the three ePRO systems sequentially at the Sir Run Run Shaw Hospital to collect patients' PRO data. RESULT: Chat-ePRO is capable of correctly generating conversation based on PRO forms (success rate: 95.7 %) and accurately extracting the PRO data instantaneously from conversation (Macro-F1: 0.95). The majority of subjective evaluations from doctors (>70 %) suggest that Chat-ePRO is able to comprehend questions and consistently generate responses. Pilot study shows that Chat-ePRO demonstrates higher response rate (9/10, 90 %) and longer interaction time (10.86 s/turn) compared to the other two methods. CONCLUSION: Our study demonstrated the feasibility of utilizing algorithms such as prompt engineering to drive LLM in completing ePRO data collection tasks, and validated that the Chat-ePRO system can effectively enhance patient compliance.

Glial growth factor 2 treatment alleviates ischemia and reperfusion-damaged integrity of the blood-brain barrier through decreasing Mfsd2a/caveolin-1-mediated transcellular and Pdlim5/YAP/TAZ-mediated paracellular permeability.

The impairment of blood-brain barrier (BBB) integrity is the pathological basis of hemorrhage transformation and vasogenic edema following thrombolysis and endovascular therapy. There is no approved drug in the clinic to reduce BBB damage after acute ischemic stroke (AIS). Glial growth factor 2 (GGF2), a recombinant version of neuregulin-1ß that can stimulates glial cell proliferation and differentiation, has been shown to alleviate free radical release from activated microglial cells. We previously found that activated microglia and proinflammatory factors could disrupt BBB after AIS. In this study we investigated the effects of GGF2 on AIS-induced BBB damage as well as the underlying mechanisms. Mouse middle cerebral artery occlusion model was established: mice received a 90-min ischemia and 22.5 h reperfusion (I/R), and were treated with GGF2 (2.5, 12.5, 50 ng/kg, i.v.) before the reperfusion. We showed that GGF2 treatment dose-dependently decreased I/R-induced BBB damage detected by Evans blue (EB) and immunoglobulin G (IgG) leakage, and tight junction protein occludin degradation. In addition, we found that GGF2 dose-dependently reversed AIS-induced upregulation of vesicular transcytosis increase, caveolin-1 (Cav-1) as well as downregulation of major facilitator superfamily domain containing 2a (Mfsd2a). Moreover, GGF2 decreased I/R-induced upregulation of PDZ and LIM domain protein 5 (Pdlim5), an adaptor protein that played an important role in BBB damage after AIS. In addition, GGF2 significantly alleviated I/R-induced reduction of YAP and TAZ, microglial cell activation and upregulation of inflammatory factors. Together, these results demonstrate that GGF2 treatment alleviates the I/R-compromised integrity of BBB by inhibiting Mfsd2a/Cav-1-mediated transcellular permeability and Pdlim5/YAP/TAZ-mediated paracellular permeability.

Isoform function prediction by Gene Ontology embedding.

MOTIVATION: High-resolution annotation of gene functions is a central task in functional genomics. Multiple proteoforms translated from alternatively spliced isoforms from a single gene are actual function performers and greatly increase the functional diversity. The specific functions of different isoforms can decipher the molecular basis of various complex diseases at a finer granularity. Multi-instance learning (MIL)-based solutions have been developed to distribute gene(bag)-level Gene Ontology (GO) annotations to isoforms(instances), but they simply presume that a particular annotation of the gene is responsible by only one isoform, neglect the hierarchical structures and semantics of massive GO terms (labels), or can only handle dozens of terms. RESULTS: We propose an efficacy approach IsofunGO to differentiate massive functions of isoforms by GO embedding. Particularly, IsofunGO first introduces an attributed hierarchical network to model massive GO terms, and a GO network embedding strategy to learn compact representations of GO terms and project GO annotations of genes into compressed ones, this strategy not only explores and preserves hierarchy between GO terms but also greatly reduces the prediction load. Next, it develops an attention-based MIL network to fuse genomics and transcriptomics data of isoforms and predict isoform functions by referring to compressed annotations. Extensive experiments on benchmark datasets demonstrate the efficacy of IsofunGO. Both the GO embedding and attention mechanism can boost the performance and interpretability. AVAILABILITYAND IMPLEMENTATION: The code of IsofunGO is available at http://www.sdu-idea.cn/codes.php?name=IsofunGO. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Intraoperative Transfusion of Autologous Blood Protects from Acute Kidney Injury after Pediatric Congenital Heart Surgery.

Background: Acute kidney injury (AKI) is a common complication after pediatric cardiac surgery. And autologous blood transfusion (ABT) is an important predictor of postoperative AKI. Unlike previous studies, which mainly focused on the correlation between ABT and AKI, the current study focuses heavily on the causal relationship between them, thus providing guidance for the treatment of patients during hospitalization to reduce the occurrence of AKI. Methods: A retrospective cohort of 3386 patients extracted from the Pediatric Intensive Care database was used for statistical analysis, multifactorial analysis, and causal inference. Characteristics that were correlated with ABT and AKI were categorized as confounders, instrumental variables, and effect modifiers, and were entered into the DoWhy causal inference model to determine causality. The calculated average treatment effect (ATE) was compared with the results of the multifactorial analysis. Results: The adjusted odds ratio (OR) for ABT volume was obtained by multifactorial analysis as 0.964. The DoWhy model refute test was able to indicate a causal relationship between ABT and AKI. Any ABT reduces AKI about 15.3%-18.8% by different estimation methods. The ATE regarding the amount of ABT was -0.0088, suggesting that every 1 mL/kg of ABT reduced the risk of AKI by 0.88%. Conclusions: Intraoperative transfusion of autologous blood can have a protective effect against postoperative AKI.

An ensemble learning with active sampling to predict the prognosis of postoperative non-small cell lung cancer patients.

BACKGROUND: Lung cancer is the leading cause of cancer death worldwide. Prognostic prediction plays a vital role in the decision-making process for postoperative non-small cell lung cancer (NSCLC) patients. However, the high imbalance ratio of prognostic data limits the development of effective prognostic prediction models. METHODS: In this study, we present a novel approach, namely ensemble learning with active sampling (ELAS), to tackle the imbalanced data problem in NSCLC prognostic prediction. ELAS first applies an active sampling mechanism to query the most informative samples to update the base classifier to give it a new perspective. This training process is repeated until no enough samples are queried. Next, an internal validation set is employed to evaluate the base classifiers, and the ones with the best performances are integrated as the ensemble model. Besides, we set up multiple initial training data seeds and internal validation sets to ensure the stability and generalization of the model. RESULTS: We verified the effectiveness of the ELAS on a real clinical dataset containing 1848 postoperative NSCLC patients. Experimental results showed that the ELAS achieved the best averaged 0.736 AUROC value and 0.453 AUPRC value for 6 prognostic tasks and obtained significant improvements in comparison with the SVM, AdaBoost, Bagging, SMOTE and TomekLinks. CONCLUSIONS: We conclude that the ELAS can effectively alleviate the imbalanced data problem in NSCLC prognostic prediction and demonstrates good potential for future postoperative NSCLC prognostic prediction.

A configurable method for clinical quality measurement through electronic health records based on openEHR and CQL.

BACKGROUND: One of the primary obstacles to measure clinical quality is the lack of configurable solutions to make computers understand and compute clinical quality indicators. The paper presents a solution that can help clinical staff develop clinical quality measurement more easily and generate the corresponding data reports and visualization by a configurable method based on openEHR and Clinical Quality Language (CQL). METHODS: First, expression logic adopted from CQL was combined with openEHR to express clinical quality indicators. Archetype binding provides the clinical information models used in expression logic, terminology binding makes the medical concepts consistent used in clinical quality artifacts and metadata is regarded as the essential component for sharing and management. Then, a systematic approach was put forward to facilitate the development of clinical quality indicators and the generation of corresponding data reports and visualization. Finally, clinical physicians were invited to test our system and give their opinions. RESULTS: With the combination of openEHR and CQL, 64 indicators from Centers for Medicare & Medicaid Services (CMS) were expressed for verification and a complicated indicator was shown as an example. 68 indicators from 17 different scenes in the local environment were also expressed and computed in our system. A platform was built to support the development of indicators in a unified way. Also, an execution engine can parse and compute these indicators. Based on a clinical data repository (CDR), indicators were used to generate data reports and visualization and shown in a dashboard. CONCLUSION: Our method is capable of expressing clinical quality indicators formally. With the computer-interpretable indicators, a systematic approach can make it more easily to define clinical indicators and generate medical data reports and visualization, and facilitate the adoption of clinical quality measurements.

Antibiotics prescription for targeted therapy of pediatric invasive pneumococcal diseases in China: a multicenter retrospective study.

BACKGROUND: Streptococcus pneumoniae (S. pneumoniae) is a major cause of bacterial meningitis, septicemia and pneumonia in children. Inappropriate choice of antibiotic can have important adverse consequences for both the individual and the community. Here, we focused on penicillin/cefotaxime non-susceptibility of S. pneumoniae and evaluated appropriateness of targeted antibiotic therapy for children with IPD (invasive pneumococcal diseases) in China. METHODS: A multicenter retrospective study was conducted in 14 hospitals from 13 provinces in China. Antibiotics prescription, clinical features and resistance patterns of IPD cases from January 2012 to December 2017 were collected. Appropriateness of targeted antibiotics therapy was assessed. RESULTS: 806 IPD cases were collected. The non-susceptibility rates of S. pneumoniae to penicillin and cefotaxime were 40.9% and 20.7% respectively in 492 non-meningitis cases, whereas those were 73.2% and 43.0% respectively in 314 meningitis cases. Carbapenems were used in 21.3% of non-meningitis cases and 42.0% of meningitis cases for targeted therapy. For 390 non-meningitis cases with isolates susceptible to cefotaxime, vancomycin and linezolid were used in 17.9% and 8.7% of cases respectively for targeted therapy. For 179 meningitis cases with isolates susceptible to cefotaxime, vancomycin and linezolid were prescribed in 55.3% and 15.6% of cases respectively. Overall, inappropriate targeted therapies were identified in 361 (44.8%) of 806 IPD cases, including 232 (28.8%) cases with inappropriate use of carbapenems, 169 (21.0%) cases with inappropriate use of vancomycin and 62 (7.7%) cases with inappropriate use of linezolid. CONCLUSIONS: Antibiotic regimens for IPD definite therapy were often excessive with extensive prescription of carbapenems, vancomycin or linezolid in China. Antimicrobial stewardship programs should be implemented to improve antimicrobial use.

Intelligent checklists improve checklist compliance in the intensive care unit: a prospective before-and-after mixed-method study.

BACKGROUND: We examined whether a context and process-sensitive 'intelligent' checklist increases compliance with best practice compared with a paper checklist during intensive care ward rounds. METHODS: We conducted a single-centre prospective before-and-after mixed-method trial in a 35 bed medical and surgical ICU. Daily ICU ward rounds were observed during two periods of 8 weeks. We compared paper checklists (control) with a dynamic (digital) clinical checklist (DCC, intervention). The primary outcome was compliance with best clinical practice, measured as the percentages of checked items and unchecked critical items. Secondary outcomes included ICU stay and the usability of digital checklists. Data are presented as median (interquartile range). RESULTS: Clinical characteristics and severity of critical illness were similar during both control and intervention periods of study. A total of 36 clinicians visited 197 patients during 352 ward rounds using the paper checklist, compared with 211 patients during 366 ward rounds using the DCC. Per ICU round, a median of 100% of items (94.4-100.0) were completed by DCC, compared with 75.1% (66.7-86.4) by paper checklist (P=0.03). No critical items remained unchecked by the DCC, compared with 15.4% (8.3-27.3) by the paper checklist (P=0.01). The DCC was associated with reduced ICU stay (1 day [1-3]), compared with the paper checklist (2 days [1-4]; P=0.05). Usability of the DCC was judged by clinicians to require further improvement. CONCLUSIONS: A digital checklist improved compliance with best clinical practice, compared with a paper checklist, during ward rounds on a mixed ICU. CLINICAL TRIAL REGISTRATION: NCT03599856.

Automatic RadLex coding of Chinese structured radiology reports based on text similarity ensemble.

BACKGROUND: Standardized coding of plays an important role in radiology reports' secondary use such as data analytics, data-driven decision support, and personalized medicine. RadLex, a standard radiological lexicon, can reduce subjective variability and improve clarity in radiology reports. RadLex coding of radiology reports is widely used in many countries, but translation and localization of RadLex in China are far from being established. Although automatic RadLex coding is a common way for non-standard radiology reports, the high-accuracy cross-language RadLex coding is hardly achieved due to the limitation of up-to-date auto-translation and text similarity algorithms and still requires further research. METHODS: We present an effective approach that combines a hybrid translation and a Multilayer Perceptron weighting text similarity ensemble algorithm for automatic RadLex coding of Chinese structured radiology reports. Firstly, a hybrid way to integrate Google neural machine translation and dictionary translation helps to optimize the translation of Chinese radiology phrases to English. The dictionary is made up of 21,863 Chinese-English radiological term pairs extracted from several free medical dictionaries. Secondly, four typical text similarity algorithms are introduced, which are Levenshtein distance, Jaccard similarity coefficient, Word2vec Continuous bag-of-words model, and WordNet Wup similarity algorithms. Lastly, the Multilayer Perceptron model has been used to synthesize the contextual, lexical, character and syntactical information of four text similarity algorithms to promote precision, in which four similarity scores of two terms are taken as input and the output presents whether the two terms are synonyms. RESULTS: The results show the effectiveness of the approach with an F1-score of 90.15%, a precision of 91.78% and a recall of 88.59%. The hybrid translation algorithm has no negative effect on the final coding, F1-score has increased by 21.44% and 8.12% compared with the GNMT algorithm and dictionary translation. Compared with the single similarity, the result of the MLP weighting similarity algorithm is satisfactory that has a 4.48% increase compared with the best single similarity algorithm, WordNet Wup. CONCLUSIONS: The paper proposed an innovative automatic cross-language RadLex coding approach to solve the standardization of Chinese structured radiology reports, that can be taken as a reference to automatic cross-language coding.

Application of openEHR archetypes to automate data quality rules for electronic health records: a case study.

BACKGROUND: Ensuring data is of appropriate quality is essential for the secondary use of electronic health records (EHRs) in research and clinical decision support. An effective method of data quality assessment (DQA) is automating data quality rules (DQRs) to replace the time-consuming, labor-intensive manual process of creating DQRs, which is difficult to guarantee standard and comparable DQA results. This paper presents a case study of automatically creating DQRs based on openEHR archetypes in a Chinese hospital to investigate the feasibility and challenges of automating DQA for EHR data. METHODS: The clinical data repository (CDR) of the Shanxi Dayi Hospital is an archetype-based relational database. Four steps are undertaken to automatically create DQRs in this CDR database. First, the keywords and features relevant to DQA of archetypes were identified via mapping them to a well-established DQA framework, Kahn's DQA framework. Second, the templates of DQRs in correspondence with these identified keywords and features were created in the structured query language (SQL). Third, the quality constraints were retrieved from archetypes. Fourth, these quality constraints were automatically converted to DQRs according to the pre-designed templates and mapping relationships of archetypes and data tables. We utilized the archetypes of the CDR to automatically create DQRs to meet quality requirements of the Chinese Application-Level Ranking Standard for EHR Systems (CARSES) and evaluated their coverage by comparing with expert-created DQRs. RESULTS: We used 27 archetypes to automatically create 359 DQRs. 319 of them are in agreement with the expert-created DQRs, covering 84.97% (311/366) requirements of the CARSES. The auto-created DQRs had varying levels of coverage of the four quality domains mandated by the CARSES: 100% (45/45) of consistency, 98.11% (208/212) of completeness, 54.02% (57/87) of conformity, and 50% (11/22) of timeliness. CONCLUSION: It's feasible to create DQRs automatically based on openEHR archetypes. This study evaluated the coverage of the auto-created DQRs to a typical DQA task of Chinese hospitals, the CARSES. The challenges of automating DQR creation were identified, such as quality requirements based on semantic, and complex constraints of multiple elements. This research can enlighten the exploration of DQR auto-creation and contribute to the automatic DQA.

A novel deep learning approach to extract Chinese clinical entities for lung cancer screening and staging.

BACKGROUND: Computed tomography (CT) reports record a large volume of valuable information about patients' conditions and the interpretations of radiology images from radiologists, which can be used for clinical decision-making and further academic study. However, the free-text nature of clinical reports is a critical barrier to use this data more effectively. In this study, we investigate a novel deep learning method to extract entities from Chinese CT reports for lung cancer screening and TNM staging. METHODS: The proposed approach presents a new named entity recognition algorithm, namely the BERT-based-BiLSTM-Transformer network (BERT-BTN) with pre-training, to extract clinical entities for lung cancer screening and staging. Specifically, instead of traditional word embedding methods, BERT is applied to learn the deep semantic representations of characters. Following the long short-term memory layer, a Transformer layer is added to capture the global dependencies between characters. Besides, pre-training technique is employed to alleviate the problem of insufficient labeled data. RESULTS: We verify the effectiveness of the proposed approach on a clinical dataset containing 359 CT reports collected from the Department of Thoracic Surgery II of Peking University Cancer Hospital. The experimental results show that the proposed approach achieves an 85.96% macro-F1 score under exact match scheme, which improves the performance by 1.38%, 1.84%, 3.81%,4.29%,5.12%,5.29% and 8.84% compared to BERT-BTN, BERT-LSTM, BERT-fine-tune, BERT-Transformer, FastText-BTN, FastText-BiLSTM and FastText-Transformer, respectively. CONCLUSIONS: In this study, we developed a novel deep learning method, i.e., BERT-BTN with pre-training, to extract the clinical entities from Chinese CT reports. The experimental results indicate that the proposed approach can efficiently recognize various clinical entities about lung cancer screening and staging, which shows the potential for further clinical decision-making and academic research.

A computational method to quantitatively measure pediatric drug safety using electronic medical records.

BACKGROUND: Drug safety in children is a major concern; however, there is still a lack of methods for quantitatively measuring, let alone to improving, drug safety in children under different clinical conditions. To assess pediatric drug safety under different clinical conditions, a computational method based on Electronic Medical Record (EMR) datasets was proposed. METHODS: In this study, a computational method was designed to extract the significant drug-diagnosis associations (based on a Bonferroni-adjusted hypergeometric P-value < 0.05) among drug and diagnosis co-occurrence in EMR datasets. This allows for differences between pediatric and adult drug use to be compared based on different EMR datasets. The drug-diagnosis associations were further used to generate drug clusters under specific clinical conditions using unsupervised clustering. A 5-layer quantitative pediatric drug safety level was proposed based on the drug safety statement of the pediatric labeling of each drug. Therefore, the drug safety levels under different pediatric clinical conditions were calculated. Two EMR datasets from a 1900-bed children's hospital and a 2000-bed general hospital were used to test this method. RESULTS: The comparison between the children's hospital and the general hospital showed unique features of pediatric drug use and identified the drug treatment gap between children and adults. In total, 591 drugs were used in the children's hospital; 18 drug clusters that were associated with certain clinical conditions were generated based on our method; and the quantitative drug safety levels of each drug cluster (under different clinical conditions) were calculated, analyzed, and visualized. CONCLUSION: With this method, quantitative drug safety levels under certain clinical conditions in pediatric patients can be evaluated and compared. If there are longitudinal data, improvements can also be measured. This method has the potential to be used in many population-level, health data-based drug safety studies.

Development of an openEHR Template for COVID-19 Based on Clinical Guidelines.

BACKGROUND: The coronavirus disease (COVID-19) was discovered in China in December 2019. It has developed into a threatening international public health emergency. With the exception of China, the number of cases continues to increase worldwide. A number of studies about disease diagnosis and treatment have been carried out, and many clinically proven effective results have been achieved. Although information technology can improve the transferring of such knowledge to clinical practice rapidly, data interoperability is still a challenge due to the heterogeneous nature of hospital information systems. This issue becomes even more serious if the knowledge for diagnosis and treatment is updated rapidly as is the case for COVID-19. An open, semantic-sharing, and collaborative-information modeling framework is needed to rapidly develop a shared data model for exchanging data among systems. openEHR is such a framework and is supported by many open software packages that help to promote information sharing and interoperability. OBJECTIVE: This study aims to develop a shared data model based on the openEHR modeling approach to improve the interoperability among systems for the diagnosis and treatment of COVID-19. METHODS: The latest Guideline of COVID-19 Diagnosis and Treatment in China was selected as the knowledge source for modeling. First, the guideline was analyzed and the data items used for diagnosis and treatment, and management were extracted. Second, the data items were classified and further organized into domain concepts with a mind map. Third, searching was executed in the international openEHR Clinical Knowledge Manager (CKM) to find the existing archetypes that could represent the concepts. New archetypes were developed for those concepts that could not be found. Fourth, these archetypes were further organized into a template using Ocean Template Editor. Fifth, a test case of data exchanging between the clinical data repository and clinical decision support system based on the template was conducted to verify the feasibility of the study. RESULTS: A total of 203 data items were extracted from the guideline in China, and 16 domain concepts (16 leaf nodes in the mind map) were organized. There were 22 archetypes used to develop the template for all data items extracted from the guideline. All of them could be found in the CKM and reused directly. The archetypes and templates were reviewed and finally released in a public project within the CKM. The test case showed that the template can facilitate the data exchange and meet the requirements of decision support. CONCLUSIONS: This study has developed the openEHR template for COVID-19 based on the latest guideline from China using openEHR modeling methodology. It represented the capability of the methodology for rapidly modeling and sharing knowledge through reusing the existing archetypes, which is especially useful in a new and fast-changing area such as with COVID-19.

Interpretable clinical prediction via attention-based neural network.

BACKGROUND: The interpretability of results predicted by the machine learning models is vital, especially in the critical fields like healthcare. With the increasingly adoption of electronic healthcare records (EHR) by the medical organizations in the last decade, which accumulated abundant electronic patient data, neural networks or deep learning techniques are gradually being applied to clinical tasks by utilizing the huge potential of EHR data. However, typical deep learning models are black-boxes, which are not transparent and the prediction outcomes of which are difficult to interpret. METHODS: To remedy this limitation, we propose an attention neural network model for interpretable clinical prediction. In detail, the proposed model employs an attention mechanism to capture critical/essential features with their attention signals on the prediction results, such that the predictions generated by the neural network model can be interpretable. RESULTS: We evaluate our proposed model on a real-world clinical dataset consisting of 736 samples to predict readmissions for heart failure patients. The performance of the proposed model achieved 66.7 and 69.1% in terms of accuracy and AUC, respectively, and outperformed the baseline models. Besides, we displayed patient-specific attention weights, which can not only help clinicians understand the prediction outcomes, but also assist them to select individualized treatment strategies or intervention plans. CONCLUSIONS: The experimental results demonstrate that the proposed model can improve both the prediction performance and interpretability by equipping the model with an attention mechanism.

Deep representation learning for individualized treatment effect estimation using electronic health records.

Utilizing clinical observational data to estimate individualized treatment effects (ITE) is a challenging task, as confounding inevitably exists in clinical data. Most of the existing models for ITE estimation tackle this problem by creating unbiased estimators of the treatment effects. Although valuable, learning a balanced representation is sometimes directly opposed to the objective of learning an effective and discriminative model for ITE estimation. We propose a novel hybrid model bridging multi-task deep learning and K-nearest neighbors (KNN) for ITE estimation. In detail, the proposed model firstly adopts multi-task deep learning to extract both outcome-predictive and treatment-specific latent representations from Electronic Health Records (EHR), by jointly performing the outcome prediction and treatment category classification. Thereafter, we estimate counterfactual outcomes by KNN based on the learned hidden representations. We validate the proposed model on a widely used semi-simulated dataset, i.e. IHDP, and a real-world clinical dataset consisting of 736 heart failure (HF) patients. The performance of our model remains robust and reaches 1.7 and 0.23 in terms of Precision in the estimation of heterogeneous effect (PEHE) and average treatment effect (ATE), respectively, on IHDP dataset, and 0.703 and 0.796 in terms of accuracy and F1 score respectively, on HF dataset. The results demonstrate that the proposed model achieves competitive performance over state-of-the-art models. In addition, the results reveal several findings which are consistent with existing medical domain knowledge, and discover certain suggestive hypotheses that could be validated through further investigations in the clinical domain.

Using the distance between sets of hierarchical taxonomic clinical concepts to measure patient similarity.

BACKGROUND: Many clinical concepts are standardized under a categorical and hierarchical taxonomy such as ICD-10, ATC, etc. These taxonomic clinical concepts provide insight into semantic meaning and similarity among clinical concepts and have been applied to patient similarity measures. However, the effects of diverse set sizes of taxonomic clinical concepts contributing to similarity at the patient level have not been well studied. METHODS: In this paper the most widely used taxonomic clinical concepts system, ICD-10, was studied as a representative taxonomy. The distance between ICD-10-coded diagnosis sets is an integrated estimation of the information content of each concept, the similarity between each pairwise concepts and the similarity between the sets of concepts. We proposed a novel method at the set-level similarity to calculate the distance between sets of hierarchical taxonomic clinical concepts to measure patient similarity. A real-world clinical dataset with ICD-10 coded diagnoses and hospital length of stay (HLOS) information was used to evaluate the performance of various algorithms and their combinations in predicting whether a patient need long-term hospitalization or not. Four subpopulation prototypes that were defined based on age and HLOS with different diagnoses set sizes were used as the target for similarity analysis. The F-score was used to evaluate the performance of different algorithms by controlling other factors. We also evaluated the effect of prototype set size on prediction precision. RESULTS: The results identified the strengths and weaknesses of different algorithms to compute information content, code-level similarity and set-level similarity under different contexts, such as set size and concept set background. The minimum weighted bipartite matching approach, which has not been fully recognized previously showed unique advantages in measuring the concepts-based patient similarity. CONCLUSIONS: This study provides a systematic benchmark evaluation of previous algorithms and novel algorithms used in taxonomic concepts-based patient similarity, and it provides the basis for selecting appropriate methods under different clinical scenarios.

Evidential MACE prediction of acute coronary syndrome using electronic health records.

BACKGROUND: Major adverse cardiac event (MACE) prediction plays a key role in providing efficient and effective treatment strategies for patients with acute coronary syndrome (ACS) during their hospitalizations. Existing prediction models have limitations to cope with imprecise and ambiguous clinical information such that clinicians cannot reach to reliable MACE prediction results for individuals. METHODS: To remedy it, this study proposes a hybrid method using Rough Set Theory (RST) and Dempster-Shafer Theory (DST) of evidence. In details, four state-of-the-art models, including one traditional ACS risk scoring model, i.e., GRACE, and three machine learning based models, i.e., Support Vector Machine, L1-Logistic Regression, and Classification and Regression Tree, are employed to generate initial MACE prediction results, and then RST is applied to determine the weights of the four single models. After that, the acquired prediction results are assumed as basic beliefs for the problem propositions and in this way, an evidential prediction result is generated based on DST in an integrative manner. RESULTS: Having applied the proposed method on a clinical dataset consisting of 2930 ACS patient samples, our model achieves 0.715 AUC value with competitive standard deviation, which is the best prediction results comparing with the four single base models and two baseline ensemble models. CONCLUSIONS: Facing with the limitations in traditional ACS risk scoring models, machine learning models and the uncertainties of EHR data, we present an ensemble approach via RST and DST to alleviate this problem. The experimental results reveal that our proposed method achieves better performance for the problem of MACE prediction when compared with the single models.

Work pattern of neurology nurses in a Chinese hospital: A time and motion study.

AIM: To investigate nursing work patterns in a neurology department in a Chinese hospital. BACKGROUND: Knowledge of nursing work patterns is important for nursing management and work design, and for the evaluation of organizational process change associated with the introduction of an innovation. METHODS: A time-and-motion method was used to observe 14 registered nurses' (RNs') work patterns in a neurology department in a Chinese hospital. RESULTS: There were 147 hr of observation. Overall, the most time-consuming activities were verbal communication (28.5%) and documentation (28.3%), followed by indirect care (14.6%) and direct care (14%). Compared to support RNs, charge RNs spent 20% more time on documentation and 11% more time on verbal communication. Two-thirds of the observed activities had a duration of less than 40 s. CONCLUSIONS: Communication, in verbal and written forms, consumed more than half of the nursing time. Conversely, nurses only spent about a quarter of their worktime on preparation for care provision and direct care provision. This reflects the significant role and resource-consuming nature of communication to provide safe and quality care. IMPLICATIONS FOR NURSING MANAGEMENT: Communication methods need to be enhanced to improve nursing productivity. This may be achieved through the introduction of more effective nursing documentation methods.