The value of T2-weighted MRI contrast ratio combined with DWI in evaluating the pathological grade of solid lung adenocarcinoma

AIM

To assess the predictive value of T2-weighted (T2W) magnetic resonance imaging (MRI) in combination with diffusion-weighted imaging (DWI) for determining the pathological grading of solid lung adenocarcinoma.

MATERIALS AND METHODS

The clinical and imaging data from 153 cases of solid lung adenocarcinoma (82 men, 71 women, mean age 63.2 years) confirmed at histopathology in The First Affiliated Hospital of Xi'an Jiaotong University from January 2017 to May 2022 were analysed retrospectively. Adenocarcinomas were classified into low-grade (G1 and G2) and high-grade (G3) groups following the 2020 pathological grading system proposed by the International Association for the Study of Lung Cancer. The T2-weighted contrast ratio (T2CR), calculated as the T2 signal intensity of the lung mass/nodule divided by the T2 signal intensity of the right rhomboid muscle was utilised. Two experienced radiologists reviewed the MRI images independently, measured the T2CR, and obtained apparent diffusion coefficient (ADC) values. The Mann-Whitney U-test was used to compare general characteristics (sex, age, maximum diameter), T2CR, and ADC values between the low-grade and high-grade groups. The non-parametric Kruskal-Wallis test determined differences in T2CR and ADC values among the five adenocarcinoma subtypes. Receiver characteristic curve (ROC) analysis, along with area under the curve (AUC) calculation, assessed the effectiveness of each parameter in distinguishing the pathological grade of lung adenocarcinoma. A Z-test was used to compare the AUC values.

RESULTS

Among the 153 patients with adenocarcinoma, 103 had low-grade adenocarcinoma, and 50 had high-grade adenocarcinoma. The agreement between T2CR and ADC observers was good (0.948 and 0.929, respectively). None of the parameters followed a normal distribution (p<0.05). The ADC value was lower in the high-grade adenocarcinoma group compared to the low-grade adenocarcinoma group (p=0.004), while the T2CR value was higher in the high-grade group (p=0.011). Statistically significant differences were observed in maximum diameter and gender between the two groups (p<0.001 and p=0.005, respectively), while no significant differences were noted in age (p=0.980). Among the five adenocarcinoma subtypes, only the lepidic and micropapillary subtypes displayed statistical differences in ADC values (p=0.047), with the remaining subtypes showing no statistical differences (p>0.05). The AUC values for distinguishing high-grade adenocarcinoma from low-grade adenocarcinoma were 0.645 for ADC and 0.627 for T2CR. Combining T2CR, ADC, sex, and maximum diameter resulted in an AUC of 0.778, sensitivity of 70%, and specificity of 75%. This combination significantly improved diagnostic efficiency compared to T2CR and ADC alone (p=0.008, z = 2.624; p=0.007, z = 2.679).

CONCLUSION

The MRI quantitative parameters are useful for distinguishing the pathological grades of solid lung adenocarcinoma, offering valuable insights for precise lung cancer treatment.

Multi-Feature Map Integrated Attention Model for Early Prediction of Type 2 Diabetes Using Irregular Health Examination Records

Type 2 diabetes (T2D) is a worldwide chronic disease that is difficult to cure and causes a heavy social burden. Early prediction of T2D can effectively identify high-risk populations and facilitate earlier implementation of appropriate preventive interventions. Various early prediction models for T2D have been proposed. However, these methods do not consider the following factors: 1) health examination records (HER) containing health information before diagnosis; 2) rating information containing clinical knowledge; and 3) local and global information of time-series features. These diagnostically relevant factors need to be considered. It is challenging due to irregular and multivariate time series. In this paper, we propose the multi-feature map integrated attention model (MFMAM) for early diabetes prediction using HER. Specifically, HER is converted into the multi-feature map to capture local and global volatility, as well as the sequence order of high-dimensional features. We concatenate rating indicators to introduce clinical knowledge. In addition, considering missing and temporal patterns, we utilize missing and time embedding to learn the complex transition of health status. We adopt attention mechanisms to capture essential features (channels) and time points (spatial). To evaluate the proposed model, we conducted experiments on real-world long-term HER. The results demonstrated that MFMAM outperformed baseline models on tasks of varying sequence lengths and prediction windows. Moreover, we applied our designs to baseline models, and their performance was considerably improved. The proposed model contributes to the short-term and long-term early prediction of T2D in individuals with varying information richness.

Longitudinal evaluation of cervical length and shear wave elastography in women with spontaneous preterm birth

OBJECTIVES

To compare longitudinal changes in cervical length (CL) and mean cervical shear wave elastography (CSWE) scores between women with singleton and twin pregnancies who experience spontaneous preterm birth (sPTB) and those who have term births (TB).

METHODS

This was a prospective longitudinal study of 1264 unselected women with singleton (n=1143) and twin (n=121) pregnancy attending a dedicated research clinic for screening of sPTB at 4 timepoints during pregnancy including 11-15+6 (visit 1), 16-20+6 (visit 2), 21-24+6 (visit 3) and 28-32+6 (visit 4) weeks of gestation. At each visit, a transvaginal ultrasound scan was conducted to measure the CL and the CSWE scores from six regions of interest (ROI) (inner, middle, and external parts of anterior and posterior lips) in the cervix. The mean of CSWE scores from the six ROIs were calculated for data analysis. Log10 transformation was applied to make the data Gaussian prior to statistical analysis. A multilevel mixed-effects analysis was performed to compare CL and CSWE longitudinally between sPTB and TB groups.

RESULTS

A total of 57 (4.99%) singleton pregnancies and 33 (27.27%) twin pregnancies were complicated with sPTB. Women with sPTB had shorter CL across gestation when controlling for history of cervical surgery, number of fetuses, gestational age at cervical assessment (GA), and the interaction between GA and sPTB. CL in the sPTB group was significantly lower than that of the TB group at 21-24+6 weeks (p=0.039) and 28-32+6 weeks (p<0.001). Twin pregnancies had significantly longer CL throughout pregnancy, compared to singleton pregnancies (coefficient=0.01864, p<0.001). Furthermore, after adjusting for maternal age, weight, height, body mass index (BMI), and GA, CSWE scores in sPTB group were significantly lower in the sPTB group across gestation, compared to the TB group (1.28265 vs 1.32832; p=0.013). However, in the individual visit analysis, CSWE scores in the sPTB group were significantly lower than that of the TB group only at 11-15+6 weeks (p=0.013). There was no difference in CSWE scores between singleton and twin pregnancies throughout pregnancy (coefficient=-0.00128, p=0.937).

CONCLUSION

Women with sPTB have shorter CL and softer cervix across gestation when compared to those with TB. In the individual visit analysis, the reduction in CL in the sPTB group occurs from late second trimester onwards, while the reduction in cervical stiffness in the sPTB group is observed primarily in the first trimester. Additionally, our study has found that CL is significantly shorter in singleton pregnancies compared to twin pregnancies, while cervical stiffness does not differ between the two types of pregnancy. Our findings indicate that the cervix tends to undergo a softening process prior to shortening in the sPTB cases This article is protected by copyright. All rights reserved.

RDmaster: A novel phenotype-oriented dialogue system supporting differential diagnosis of rare disease

BACKGROUND

Clinicians often lack the necessary expertise to differentially diagnose multiple underlying rare diseases (RDs) due to their complex and overlapping clinical features, leading to misdiagnoses and delayed treatments. The aim of this study is to develop a novel electronic differential diagnostic support system for RDs.

METHOD

Through integrating two Bayesian diagnostic methods, a candidate list was generated with enhance clinical interpretability for the further Q&A based differential diagnosis (DDX). To achieve an efficient Q&A dialogue strategy, we introduce a novel metric named the adaptive information gain and Gini index (AIGGI) to evaluate the expected gain of interrogated phenotypes within real-time diagnostic states.

RESULTS

This DDX tool called RDmaster has been implemented as a web-based platform (http://rdmaster.nbscn.org/). A diagnostic trial involving 238 published RD patients revealed that RDmaster outperformed existing RD diagnostic tools, as well as ChatGPT, and was shown to enhance the diagnostic accuracy through its Q&A system.

CONCLUSIONS

The RDmaster offers an effective multi-omics differential diagnostic technique and outperforms existing tools and popular large language models, particularly enhancing differential diagnosis in collecting diagnostically beneficial phenotypes.

Improved transcranial plane-wave imaging with Learned Speed-of-Sound Maps

Although transcranial ultrasound plane-wave imaging (PWI) has promising clinical application prospects, studies have shown that variable speed-of-sound (SoS) would seriously damage the quality of ultrasound images. The mismatch between the conventional constant velocity assumption and the actual SoS distribution leads to the general blurring of ultrasound images. The optimization scheme for reconstructing transcranial ultrasound image is often solved using iterative methods like full-waveform inversion. These iterative methods are computationally expensive and based on prior magnetic resonance imaging (MRI) or computed tomography (CT) information. In contrast, the multi-stencils fast marching (MSFM) method can produce accurate time travel maps for the skull with heterogeneous acoustic speed. In this study, we first propose a convolutional neural network (CNN) to predict SoS maps of the skull from PWI channel data. Then, use these maps to correct the travel time to reduce transcranial aberration. To validate the performance of the proposed method, numerical and phantom studies were conducted using a linear array transducer (L11-5v, 128 elements, pitch = 0.3 mm). Numerical simulations demonstrate that for point targets, the lateral resolution of MSFM-restored images increased by 65%, and the center position shift decreased by 89%. For the cyst targets, the eccentricity of the fitting ellipse decreased by 75%, and the center position shift decreased by 58%. In the phantom study, the lateral resolution of MSFM-restored images was increased by 49%, and the position shift was reduced by 1.72 mm. This pipeline, termed AutoSoS, thus shows the potential to correct distortions in real-time transcranial ultrasound imaging.

A Patient Similarity Network (CHDmap) to Predict Outcomes After Congenital Heart Surgery: Development and Validation Study

Background

Although evidence-based medicine proposes personalized care that considers the best evidence, it still fails to address personal treatment in many real clinical scenarios where the complexity of the situation makes none of the available evidence applicable. "Medicine-based evidence" (MBE), in which big data and machine learning techniques are embraced to derive treatment responses from appropriately matched patients in real-world clinical practice, was proposed. However, many challenges remain in translating this conceptual framework into practice.

Objective

This study aimed to technically translate the MBE conceptual framework into practice and evaluate its performance in providing general decision support services for outcomes after congenital heart disease (CHD) surgery.

Methods

Data from 4774 CHD surgeries were collected. A total of 66 indicators and all diagnoses were extracted from each echocardiographic report using natural language processing technology. Combined with some basic clinical and surgical information, the distances between each patient were measured by a series of calculation formulas. Inspired by structure-mapping theory, the fusion of distances between different dimensions can be modulated by clinical experts. In addition to supporting direct analogical reasoning, a machine learning model can be constructed based on similar patients to provide personalized prediction. A user-operable patient similarity network (PSN) of CHD called CHDmap was proposed and developed to provide general decision support services based on the MBE approach.

Results

Using 256 CHD cases, CHDmap was evaluated on 2 different types of postoperative prognostic prediction tasks: a binary classification task to predict postoperative complications and a multiple classification task to predict mechanical ventilation duration. A simple poll of the k-most similar patients provided by the PSN can achieve better prediction results than the average performance of 3 clinicians. Constructing logistic regression models for prediction using similar patients obtained from the PSN can further improve the performance of the 2 tasks (best area under the receiver operating characteristic curve=0.810 and 0.926, respectively). With the support of CHDmap, clinicians substantially improved their predictive capabilities.

Conclusions

Without individual optimization, CHDmap demonstrates competitive performance compared to clinical experts. In addition, CHDmap has the advantage of enabling clinicians to use their superior cognitive abilities in conjunction with it to make decisions that are sometimes even superior to those made using artificial intelligence models. The MBE approach can be embraced in clinical practice, and its full potential can be realized.

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.

Recent advances in the management of postmenopausal women with non-atypical endometrial hyperplasia

Non-atypical endometrial hyperplasia is a benign disease without significant somatic genetic changes. Postmenopausal women with non-atypical endometrial hyperplasia have a significant risk of progression to endometrial cancer and persistent endometrial hyperplasia. Most cases of atypical endometrial hyperplasia in postmenopausal women are treated surgically, including hysterectomy. At present, the treatment of postmenopausal women with non-atypical endometrial hyperplasia is still controversial. Correct and timely diagnosis and treatment are of great significance to prevent progression of the lesion. This study mainly provides an updated synthesis of the literature that investigates the etiology, diagnosis and treatment of postmenopausal women with non-atypical endometrial hyperplasia. As of December 2022, a literature search related to postmenopausal non-atypical endometrial hyperplasia was conducted on the PubMed database. For most postmenopausal patients with non-atypical endometrial hyperplasia, regular re-examination should be performed during conservative treatment. For postmenopausal patients with endometrial cancer risk factors, persistent non-atypical endometrial hyperplasia or progesterone contraindications, hysterectomy and bilateral salpingo-oophorectomy should be the first choice.

Transcranial Acoustic Metamaterial Parameters Inverse Designed by Neural Networks

Objective: The objective of this work is to investigate the mapping relationship between transcranial ultrasound image quality and transcranial acoustic metamaterial parameters using inverse design methods. Impact Statement: Our study provides insights into inverse design methods and opens the route to guide the preparation of transcranial acoustic metamaterials. Introduction: The development of acoustic metamaterials has enabled the exploration of cranial ultrasound, and it has been found that the influence of the skull distortion layer on acoustic waves can be effectively eliminated by adjusting the parameters of the acoustic metamaterial. However, the interaction mechanism between transcranial ultrasound images and transcranial acoustic metamaterial parameters is unknown. Methods: In this study, 1,456 transcranial ultrasound image datasets were used to explore the mapping relationship between the quality of transcranial ultrasound images and the parameters of transcranial acoustic metamaterials. Results: The multioutput parameter prediction model of transcranial metamaterials based on deep back-propagation neural network was built, and metamaterial parameters under transcranial image evaluation indices are predicted using the prediction model. Conclusion: This inverse big data design approach paves the way for guiding the preparation of transcranial metamaterials.

[Evaluation of the efficacy and safety of Nocardia rubra cell wall skeleton immunotherapy for cervical high-risk HPV persistent infection]

Objective: To evaluate the efficacy and safety of Nocardia rubra cell wall skeleton (Nr-CWS) in the treatment of persistent cervical high-risk human papillomavirus (HR-HPV) infection. Methods: A randomized, double blind, multi-center trial was conducted. A total of 688 patients with clinically and pathologically confirmed HR-HPV infection of the cervix diagnosed in 13 hispital nationwide were recruited and divided into: (1) patients with simple HR-HPV infection lasting for 12 months or more; (2) patients with cervical intraepithelial neoplasia (CIN) Ⅰ and HR-HPV infection lasting for 12 months or more; (3) patients with the same HR-HPV subtype with no CINⅡ and more lesions after treatment with CINⅡ or CIN Ⅲ (CINⅡ/CIN Ⅲ). All participants were randomly divided into the test group and the control group at a ratio of 2∶1. The test group was locally treated with Nr-CWS freeze-dried powder and the control group was treated with freeze-dried powder without Nr-CWS. The efficacy and negative conversion rate of various subtypes of HR-HPV were evaluated at 1, 4, 8, and 12 months after treatment. The safety indicators of initial diagnosis and treatment were observed. Results: (1) This study included 555 patients with HR-HPV infection in the cervix (included 368 in the test group and 187 in the control group), with an age of (44.1±10.0) years. The baseline characteristics of the two groups of subjects, including age, proportion of Han people, weight, composition of HR-HPV subtypes, and proportion of each subgroup, were compared with no statistically significant differences (all P>0.05). (2) After 12 months of treatment, the effective rates of the test group and the control group were 91.0% (335/368) and 44.9% (84/187), respectively. The difference between the two groups was statistically significant (χ²=142.520, P<0.001). After 12 months of treatment, the negative conversion rates of HPV 16, 18, 52, and 58 infection in the test group were 79.2% (84/106), 73.3% (22/30), 83.1% (54/65), and 77.4% (48/62), respectively. The control group were 21.6% (11/51), 1/9, 35.1% (13/37), and 20.0% (8/40), respectively. The differences between the two groups were statistically significant (all P<0.001). (3) There were no statistically significant differences in vital signs (body weight, body temperature, respiration, pulse rate, systolic blood pressure, diastolic blood pressure, etc.) and laboratory routine indicators (blood cell analysis, urine routine examination) between the test group and the control group before treatment and at 1, 4, 8, and 12 months after treatment (all P>0.05); there was no statistically significant difference in the incidence of adverse reactions related to the investigational drug between the two groups of subjects [8.7% (32/368) vs 8.0% (15/187), respectively; χ²=0.073, P=0.787]. Conclusion: External use of Nr-CWS has good efficacy and safety in the treatment of high-risk HPV persistent infection in the cervix.

[Development and validation of prognostic nomogram for malignant pleural mesothelioma]

Objective: To development the prognostic nomogram for malignant pleural mesothelioma (MPM). Methods: Two hundred and ten patients pathologically confirmed as MPM were enrolled in this retrospective study from 2007 to 2020 in the People's Hospital of Chuxiong Yi Autonomous Prefecture, the First and Third Affiliated Hospital of Kunming Medical University, and divided into training (n=112) and test (n=98) sets according to the admission time. The observation factors included demography, symptoms, history, clinical score and stage, blood cell and biochemistry, tumor markers, pathology and treatment. The Cox proportional risk model was used to analyze the prognostic factors of 112 patients in the training set. According to the results of multivariate Cox regression analysis, the prognostic prediction nomogram was established. C-Index and calibration curve were used to evaluate the model's discrimination and consistency in raining and test sets, respectively. Patients were stratified according to the median risk score of nomogram in the training set. Log rank test was performed to compare the survival differences between the high and low risk groups in the two sets. Results: The median overall survival (OS) of 210 MPM patients was 384 days (IQR=472 days), and the 6-month, 1-year, 2-year, and 3-year survival rates were 75.7%, 52.6%, 19.7%, and 13.0%, respectively. Cox multivariate regression analysis showed that residence (HR=2.127, 95% CI: 1.154-3.920), serum albumin (HR=1.583, 95% CI: 1.017-2.464), clinical stage (stage Ⅳ: HR=3.073, 95% CI: 1.366-6.910) and the chemotherapy (HR=0.476, 95% CI: 0.292-0.777) were independent prognostic factors for MPM patients. The C-index of the nomogram established based on the results of Cox multivariate regression analysis in the training and test sets were 0.662 and 0.613, respectively. Calibration curves for both the training and test sets showed moderate consistency between the predicted and actual survival probabilities of MPM patients at 6 months, 1 year, and 2 years. The low-risk group had better outcomes than the high-risk group in both training (P=0.001) and test (P=0.003) sets. Conclusion: The survival prediction nomogram established based on routine clinical indicators of MPM patients provides a reliable tool for prognostic prediction and risk stratification.

Systematic exploration of eczema-associated paediatric diseases in a Chinese population of millions: A retrospective observation study

BACKGROUND

Eczema is the most common form of dermatitis and also the starting point of atopic march. Although many eczema-associated allergic and immunologic disorders have been studied, there remains a gap in the systematic quantitative knowledge regarding the relationships between all childhood disorders and eczema. This study aimed to systematically explore eczema-associated childhood diseases using a real-world, long-term clinical dataset generated from millions of children in China.

METHODS

Data were collected at 8,907,735 outpatient healthcare visits from 2,592,147 children between January 1, 2013, and August 15, 2019, at the largest comprehensive pediatric medical center in Zhejiang Province of China. The period prevalence differences in various pediatric diseases between children with and without eczema were used to test the independence of various pediatric disorders and eczema using Fisher's exact test. Bonferroni correction was used to adjust the p value in multiple testing. Odds ratio >2 with 95% confidence interval not including 1 and adjusted p < 0.05 was used to identify eczema-associated diseases.

RESULTS

Overall, 234 pediatric disorders were identified from more than 6000 different pediatric disorders. An interactive eczema-associated disease map that has related quantitative epidemiological features called ADmap was published at http://pedmap.nbscn.org/admap. Thirty-six of these disease associations have not been reported in previous studies.

CONCLUSION

This systematic exploratory study confirmed the associations of many well-known diseases with eczema in Chinese children and also identified some novel and interesting associations. These results are valuable for the development of a comprehensive approach to the management of eczema in childhood.

A Robust Phenotype-driven Likelihood Ratio Analysis Approach Assisting Interpretable Clinical Diagnosis of Rare Diseases

Phenotype-based prioritization of candidate genes and diseases has become a well-established approach for multi-omics diagnostics of rare diseases. Most current algorithms exploit semantic analysis and probabilistic statistics based on Human Phenotype Ontology and are commonly superior to naive search methods. However, these algorithms are mostly less interpretable and do not perform well in real clinical scenarios due to noise and imprecision of query terms, and the fact that individuals may not display all phenotypes of the disease they belong to. We present a Phenotype-driven Likelihood Ratio analysis approach (PheLR) assisting interpretable clinical diagnosis of rare diseases. With a likelihood ratio paradigm, PheLR estimates the posterior probability of candidate diseases and how much a phenotypic feature contributes to the prioritization result. Benchmarked using simulated and realistic patients, PheLR shows significant advantages over current approaches and is robust to noise and inaccuracy. To facilitate clinical practice and visualized differential diagnosis, PheLR is implemented as an online web tool (http://phelr.nbscn.org).

Automatic ICD-10 coding: Deep semantic matching based on analogical reasoning

Background

ICD-10 has been widely used in statistical analysis of mortality rates and medical reimbursement. Automatic ICD-10 coding is desperately needed because manually assigning codes is expensive, time-consuming, and labor-intensive. Diagnoses described in medical records differ significantly from those used in ICD-10 classification, making it impossible for existing automatic coding techniques to perform well enough to support medical billing, resource allocation, and research requirements. Meanwhile, most of the current automatic coding approaches are oriented toward English ICD-10. This method for automatically assigning ICD-10 codes to diagnoses extracted from Chinese discharge records was provided in this paper.

Method

First, BERT creates word representations of the two texts. Second, the context representation layer incorporates contextual information into the representation of each time step of the word representations using a bidirectional Long Short-Term Memory. Third, the matching layer compares each contextual embedding of the uncoded diagnosis record against a weighted version of all contextual character embeddings of the manually coded diagnosis record. The matching strategy is element-wise subtraction and element-wise multiplication and then through a neural network layer. Fourth, the matching vectors are combined using a one-layer convolutional neural network. A sigmoid is then used to output matching results.

Results

To evaluate the proposed method, 1,003,558 manually coded primary diagnoses were gathered from the homepage of the discharge medical records. The experimental results showed that the proposed method outperformed popular deep semantic matching algorithms, such as DSSM, ConvNet, ESIM, and ABCNN, and demonstrated state-of-the-art results in a single text matching with an accuracy of 0.986, a precision of 0.979, a recall of 0.983, and an F1-score of 0.981.

Conclusion

The automatic ICD-10 coding of Chinese diagnoses is successful when using the proposed deep semantic matching approach based on analogical reasoning.

[Comparative analysis of clinical diagnosis application of two intrauterine adhesion scoring criteria]

Objective: To explore the similarities and differences of China Society of Gynecology Endoscopy (CSGE) and American Fertility Society (AFS) intrauterine adhesion (IUA) scoring criteria on IUA grading and their predictive value of reproductive prognosis. Methods: From January 2016 to January 2019, a total of 1 249 patients were diagnosed with IUA by hysteroscopy at Beijing Obstetrics and Gynecology Hospital. Totally, 378 patients with complete clinical data were enrolled, and the results diagnosed by CSGT and AFS scoring criteria were compared and analyzed.And follow-up for 2 years, the pregnancy rate and live birth rate were statistical analysis. Results: (1) The grade of IUA according to AFS and CSGE scoring criteria was less consistent (κ=0.295, P<0.001). Compared with AFS, the proportion of severe IUA cases diagnosed by CSGE was significantly lower [45.8% (173/378) vs 15.1% (57/378); RR=0.22, 95%CI: 0.15-0.30, P<0.01); the proportions of both mild and moderate IUA cases were significantly higher (RR=4.16, 95%CI: 2.38-7.14; RR=2.38, 95%CI: 1.75-3.23; both P<0.01). (2) The pregnancy rates of mild, moderate and severe IUA diagnosed according to CSGE were 11/13, 64.5% (147/228), 31.8% (7/22), live birth rates were 11/13, 54.8% (125/228) and 22.7% (5/22), respectively; there were statistically significant differences between the groups (all P<0.01). The pregnancy rates of mild, moderate and severe IUA diagnosed based on AFS were 3/3, 66.9% (97/145) and 56.5% (65/115), respectively, with no statistically significant difference between the groups (P>0.05). (3) IUA grades based on both CSGE and AFS criteria were significantly negatively correlated with pregnancy rates and live birth rates (CSGE: r=-0.210, r=-0.226; AFS: r=-0.130, r=-0.147; all P<0.05). Univariate logistic regression analysis showed that CSGE had higher OR for both pregnancy rates and live birth rates compared to AFS (3.889 vs 1.657, 3.983 vs 1.554, respectrvely). Conclusions: Compared with AFS, the IUA grade based on CSGE is better related with reproductive prognosis, suggesting that the CSGE standard might be more objective and comprehensive and has better predictive value for reproductive prognosis, thus avoiding overdiagnosis and overtreatment.

Dealing With Missing, Imbalanced, and Sparse Features During the Development of a Prediction Model for Sudden Death Using Emergency Medicine Data: Machine Learning Approach

BACKGROUND

In emergency departments (EDs), early diagnosis and timely rescue, which are supported by prediction modes using ED data, can increase patients' chances of survival. Unfortunately, ED data usually contain missing, imbalanced, and sparse features, which makes it challenging to build early identification models for diseases.

OBJECTIVE

This study aims to propose a systematic approach to deal with the problems of missing, imbalanced, and sparse features for developing sudden-death prediction models using emergency medicine (or ED) data.

METHODS

We proposed a 3-step approach to deal with data quality issues: a random forest (RF) for missing values, k-means for imbalanced data, and principal component analysis (PCA) for sparse features. For continuous and discrete variables, the decision coefficient R² and the κ coefficient were used to evaluate performance, respectively. The area under the receiver operating characteristic curve (AUROC) and the area under the precision-recall curve (AUPRC) were used to estimate the model's performance. To further evaluate the proposed approach, we carried out a case study using an ED data set obtained from the Hainan Hospital of Chinese PLA General Hospital. A logistic regression (LR) prediction model for patient condition worsening was built.

RESULTS

A total of 1085 patients with rescue records and 17,959 patients without rescue records were selected and significantly imbalanced. We extracted 275, 402, and 891 variables from laboratory tests, medications, and diagnosis, respectively. After data preprocessing, the median R² of the RF continuous variable interpolation was 0.623 (IQR 0.647), and the median of the κ coefficient for discrete variable interpolation was 0.444 (IQR 0.285). The LR model constructed using the initial diagnostic data showed poor performance and variable separation, which was reflected in the abnormally high odds ratio (OR) values of the 2 variables of cardiac arrest and respiratory arrest (201568034532 and 1211118945, respectively) and an abnormal 95% CI. Using processed data, the recall of the model reached 0.746, the F₁-score was 0.73, and the AUROC was 0.708.

CONCLUSIONS

The proposed systematic approach is valid for building a prediction model for emergency patients.

A time-aware attention model for prediction of acute kidney injury after pediatric cardiac surgery

OBJECTIVE

Acute kidney injury (AKI) is a common complication after pediatric cardiac surgery, and the early detection of AKI may allow for timely preventive or therapeutic measures. However, current AKI prediction researches pay less attention to time information among time-series clinical data and model building strategies that meet complex clinical application scenario. This study aims to develop and validate a model for predicting postoperative AKI that operates sequentially over individual time-series clinical data.

MATERIALS AND METHODS

A retrospective cohort of 3386 pediatric patients extracted from PIC database was used for training, calibrating, and testing purposes. A time-aware deep learning model was developed and evaluated from 3 clinical perspectives that use different data collection windows and prediction windows to answer different AKI prediction questions encountered in clinical practice. We compared our model with existing state-of-the-art models from 3 clinical perspectives using the area under the receiver operating characteristic curve (ROC AUC) and the area under the precision-recall curve (PR AUC).

RESULTS

Our proposed model significantly outperformed the existing state-of-the-art models with an improved average performance for any AKI prediction from the 3 evaluation perspectives. This model predicted 91% of all AKI episodes using data collected at 24 h after surgery, resulting in a ROC AUC of 0.908 and a PR AUC of 0.898. On average, our model predicted 83% of all AKI episodes that occurred within the different time windows in the 3 evaluation perspectives. The calibration performance of the proposed model was substantially higher than the existing state-of-the-art models.

CONCLUSIONS

This study showed that a deep learning model can accurately predict postoperative AKI using perioperative time-series data. It has the potential to be integrated into real-time clinical decision support systems to support postoperative care planning.

DNA methylation and waist-to-hip ratio: an epigenome-wide association study in Chinese monozygotic twins

PURPOSE

Epigenetic signatures such as DNA methylation may be associated with specific obesity traits. We performed an epigenome-wide association study (EWAS) by combining with the waist-to-hip ratio (WHR)-discordant monozygotic (MZ) twin design in an attempt to identify genetically independent DNA methylation marks associated with abdominal obesity in Northern Han Chinese and to determine the causation underlying.

METHODS

A total of 60 WHR discordant MZ twin pairs were selected from the Qingdao Twin Registry, China. Generalized estimated equation (GEE) model was used to regress the methylation level of CpG sites on WHR. The Inference about Causation through Examination of FAmiliaL CONfounding (ICE FALCON) was used to assess the temporal relationship between methylation and WHR. Gene expression analysis was conducted to validate the results of differentially methylated analyses.

RESULTS

EWAS identified 92 CpG sites with the level of P < 10^- 4 which were annotated to 32 genes, especially CADPS2, TUSC5, ZCCHC14, CORO7, COL23A1, CACNA1C, CYP26B1, and BCAT1. ICE FALCON showed significant causality between DNA methylation of several genes and WHR (P < 0.05). In region-based analysis, 14 differentially methylated regions (DMRs) located at 15 genes (slk-corrected P < 0.05) were detected. The gene expression analysis identified the significant correlation between expression levels of 5 differentially methylated genes and WHR (P < 0.05).

CONCLUSIONS

Our study identifies the associations between specific epigenetic variations and WHR in Northern Han Chinese. These DNA methylation signatures may have value as diagnostic biomarkers and provide novel insights into the molecular mechanisms of pathogenesis.

Deep Dynamic Patient Similarity Analysis: Model Development and Validation in ICU

BACKGROUND

Personalized medicine requires the patient similarity analysis for providing specific treatments tailed for each patient. However, the patient similarity analysis in personalized clinical scenarios encounters challenges, which are twofold. First, heterogeneous and multi-type data are usually recorded to Electronic Health Records (EHRs) during the course of admissions, which makes it difficult to measure the patient similarity. Second, disease progression manifests diverse disease states at different times, which brings sequential complexity to dynamically retrieve similar patients' sequences.

MATERIALS AND METHODS

To overcome the above-mentioned challenges, we propose a novel dynamic patient similarity analysis model based on deep learning. Specifically, the proposed model embeds the multi-type and heterogeneous data into hidden representations with a specially designed embedding and attention module. Thereafter, the proposed model retrieves similar patients' sequences based on these hidden representations in a dynamic manner. More importantly, we adopt two clinical tasks, i.e., diagnosis prediction and medication recommendation, to validate the effectiveness of the proposed model. It is worth noticing that the proposed model integrates a drug-drug interaction (DDI) knowledge graph in the medication recommendation task to reduce adverse reactions caused by combinational treatments, such that a more rational strategy can be realized. We evaluate our proposed model using the critical care database MIMIC-III, which includes 5,430 patients covering 14,096 clinical visits.

RESULTS

The proposed model outperforms several state-of-the-art methods. For diagnosis prediction, the average PR-AUC score of the proposed model reaches 0.6200, which is significantly higher than that of the baseline models (0.2497∼0.5407). Meanwhile, for medication recommendation, the average PR-AUC of the proposed model is 0.6682 (Jaccard: 0.4070; F1: 0.5672; Recall: 0.7832) whereas the K-nearest model can only reach 0.3805 (Jaccard: 0.3911; F1: 0.5465; Recall: 0.5705). In addition, our proposed model achieves a lower DDI rate.

CONCLUSION

We propose a novel dynamic patient similarity analysis model, which can be implemented into a decision support system for clinical tasks including diagnosis prediction, surgical procedure selection, medication recommendation, etc. Also, the proposed model serves as an explainable protocol in clinical practice thanks to its analogy to real clinical reasoning where a doctor diagnoses diseases and prescribes medications according to the previous cured patients empirically.

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.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12911-022-01960-0.

P11.75.B Survival benefit of radiotherapy and surgery in patients with lung cancer brain metastases with poor prognosis factors

Background

Radiotherapy and surgery are the standard treatments for lung cancer brain metastases (BMs). However, limitted studies focused on the treatments for patients with lung cancer BMs with poor prognosis factors. The purpose of this study was to investigate the effects of radiotherapy and surgery in patients with lung cancer BMs with poor prognosis factors, providing reference for clinical strategies.

Material and Methods

We analyzed retrospectively 714 patients with lung cancer BMs. A 1:1 propensity score matching (PSM) was performed to balance potential confounders. Analyses of overall survival (OS) and risk factors for OS were assessed by log-rank test and Cox proportional hazard model.

Results

Age ≥65 years, Karnofsky Performance Scale (KPS) score ≤70, anaplastic large-cell lymphoma kinase (ALK)/epidermal growth factor receptor (EGFR) wild type, extracranial metastases, non-surgery and non-radiotherapy led to poor prognosis. Patients were stratified according to these factors. Radiotherapy and surgery showed no survival benefit in patients with aged ≥65 years or pretreatment KPS score ≤70 before and after PSM. Before PSM, whole brain radiotherapy (WBRT) improved the OS and predicted good prognosis in patients with ALK/EGFR wild type or extracranial metastases. WBRT also predicted good prognosis in patients with non-surgery. Stereotactic radiosurgery (SRS) improved the OS and predicted good prognosis in patients with ALK/EGFR wild type or non-surgery. WBRT plus SRS improved the OS and predicted good prognosis in patients with extracranial metastases or non-surgery. WBRT plus SRS also predicted good prognosis in patients with ALK/EGFR wild type. Surgery improved the OS and predicted good prognosis in patients with non-radiotherapy. After PSM, SRS improved the OS and predicted good prognosis in patients with non-surgery. WBRT plus SRS improved the OS and predicted good prognosis in patients with non-surgery or extracranial metastases. WBRT plus SRS also predicted good prognosis in patients with ALK/EGFR wild type. Surgery improved the OS of patients with non-radiotherapy. We defined that the treatment would provide significant survival benefit if it both prolonged the OS and predicted good prognosis. Meanwhile, the results after PSM were more convincing than the results before PSM.

Conclusion

Radiotherapy has significant survival benefit in patients with lung cancer BMs with poor prognosis factors, including patients with ALK/EGFR wild type or extracranial metastases or non-surgery. Surgery only has significant survival benefit in patients with non-radiotherapy.

SNPMap—An integrated visual SNP interpretation tool

New technologies, such as next-generation sequencing, have advanced the ability to diagnose diseases and improve prognosis but require the identification of thousands of variants in each report based on several databases scattered across places. Curating an integrated interpretation database is time-consuming, costly, and needs regular update. On the other hand, the automatic curation of knowledge sources always results in overloaded information. In this study, an automated pipeline was proposed to create an integrated visual single-nucleotide polymorphism (SNP) interpretation tool called SNPMap. SNPMap pipelines periodically obtained SNP-related information from LitVar, PubTator, and GWAS Catalog API tools and presented it to the user after extraction, integration, and visualization. Keywords and their semantic relations to each SNP are rendered into two graphs, with their significance represented by the size/width of circles/lines. Moreover, the most related SNPs for each keyword that appeared in SNPMap were calculated and sorted. SNPMap retains the advantage of an automatic process while assisting users in accessing more lucid and detailed information through visualization and integration with other materials.

MRI-based radiomics analysis in differentiating solid non-small-cell from small-cell lung carcinoma: a pilot study

AIM

To investigate the ability of a T2-weighted (W) magnetic resonance imaging (MRI)-based radiomics signature to differentiate solid non-small-cell lung carcinoma (NSCLC) from small-cell lung carcinoma (SCLC).

MATERIALS AND METHODS

The present retrospective study enrolled 152 eligible patients (NSCLC = 125, SCLC = 27). All patients underwent MRI using a 3 T scanner and radiomics features were extracted from T2W MRI. The least absolute shrinkage and selection operator (LASSO) logistic regression model was used to identify the optimal radiomics features for the construction of a radiomics model to differentiate solid NSCLC from SCLC. Threefold cross validation repeated 10 times was used for model training and evaluation. The conventional MRI morphology features of the lesions were also evaluated. The performance of the conventional MRI morphological features, and the radiomics signature model and nomogram model (combining radiomics signature with conventional MRI morphological features) was evaluated using receiver operating characteristic (ROC) curve analysis.

RESULTS

Five optimal features were chosen to build a radiomics signature. There was no significant difference in age, gender, and the largest diameter. The radiomics signature and conventional MRI morphological features (only pleural indentation and lymph node enlargement) were independent predictive factors for differentiating solid NSCLC from SCLC. The area under the ROC curves (AUCs) for MRI morphological features, and the radiomics model, and nomogram model was 0.69, 0.85, and 0.90 (ROC), respectively.

CONCLUSIONS

The T2W MRI-based radiomics signature is a potential non-invasive approach for distinguishing solid NSCLC from SCLC.

Making Specific Plan Improves Physical Activity and Healthy Eating for Community-Dwelling Patients With Chronic Conditions: A Systematic Review and Meta-Analysis

Background

Implementation intention formed by making a specific action plan has been proved effective in improving physical activity (PA) and dietary behavior (DB) for the general, healthy population, but there has been no meta-analysis of their effectiveness for patients with chronic conditions. This research aims to analyze several explanatory factors and overall effect of implementation intention on behavioral and health-related outcomes among community-dwelling patients.

Methods

We searched CIHNAL (EBSCO), PUBMED, Web of Science, Science Direct, SAGE Online, Springer Link, Taylor & Francis, Scopus, Wiley Online Library, CNKI, and five other databases for eligible studies. Random-effects meta-analysis was conducted to estimate effect sizes of implementation intention on outcomes, including PA, DB, weight, and body mass index. And the eligible studies were assessed by the Cochrane Collaboration's tool for risk of bias assessment. Sensitivity analysis adopted sequential algorithm and the p-curve analysis method.

Results

A total of 54 studies were identified. Significant small effect sizes of the intervention were found for PA [standard mean difference (SMD) 0.24, 95% confidence interval (CI) (0.10, 0.39)] and for the DB outcome [SMD -0.25, 95% CI (-0.34, -0.15)]. In moderation analysis, the intervention was more effective in improving PA for men (p < 0.001), older adults (p = 0.006), and obese/overweight patients with complications (p = 0.048) and when the intervention was delivered by a healthcare provider (p = 0.01).

Conclusion

Implementation intentions are effective in improving PA and DB for community dwelling patients with chronic conditions. The review provides evidence to support the future application of implementation intention intervention. Besides, the findings from this review offer different directions to enhance the effectiveness of this brief and potential intervention in improving patients' PA and DB.

Systematic Review Registration

https://www.crd.york.ac.uk/PROSPERO/display_record.php?RecordID=160491.

Ultrasonic Methods for Brain Imaging: Techniques and Implications

Brain imaging technology is widely used in the diagnosis of brain diseases. Computed tomography and magnetic resonance imaging are the most common imaging modalities used for clinical brain imaging, whereas ultrasound is rarely used because the skull substantially reduces the incident energy of ultrasonic waves to levels too low for imaging. However, remarkable developments of novel technologies in ultrasound brain imaging have been achieved recently, including Doppler-based imaging, contrast agent imaging, ultrasound elastography, and phase compensation imaging. Doppler-based imaging, including ultrafast Doppler imaging and functional ultrasound, is able to obtain reliable blood flow information and has the best penetration depth and highest temporal resolution. Contrast agent brain imaging, including ultrasound localization microscopy, has the best spatial resolution, and its temporal resolution can be maintained within a few milliseconds. Ultrasound elastography reflects the stiffness of brain tissues. Phase correction imaging, such as time reversal mirror and spatiotemporal inverse filter, aims at focusing smoothly in the skull. These methods have been widely performed on animal models, newborn children, and adults in preclinical studies, with results demonstrating great potential in the diagnosis and treatment of brain diseases. This review discusses the ultrasound methods developed in recent years for brain imaging and highlights the promising future they hold.

Risk factors for central venous catheter-associated deep venous thrombosis in pediatric critical care settings identified by fusion model

Background

An increase in the incidence of central venous catheter (CVC)-related thrombosis (CRT) has been reported in pediatric intensive care patients over the past decade. Risk factors for the development of CRT are not well understood, especially in children. The study objective was to identify potential clinical risk factors associated with CRT with novel fusion machine learning models.

Methods

Patients aged 0–18 who were admitted to intensive care units from December 2015 to December 2018 and underwent at least one CVC placement were included. Two fusion model approaches (stacking and blending) were used to build a better performance model based on three widely used machine learning models (logistic regression, random forest and gradient boosting decision tree). High-impact risk factors were identified based on their contribution in both fusion artificial intelligence models.

Results

A total of 478 factors of 3871 patients and 3927 lines were used to build fusion models, one of which achieved quite satisfactory performance (AUC = 0.82, recall = 0.85, accuracy = 0.65) in 5-fold cross validation. A total of 11 risk factors were identified based on their independent contributions to the two fusion models. Some risk factors, such as D-dimer, thrombin time, blood acid-base balance-related factors, dehydrating agents, lymphocytes and basophils were identified or confirmed to play an important role in CRT in children.

Conclusions

The fusion model, which achieves better performance in CRT prediction, can better understand the risk factors for CRT and provide potential biomarkers and measures for thromboprophylaxis in pediatric intensive care settings.

Supplementary information

The online version contains supplementary material available at 10.1186/s12959-022-00378-y.

Multi-label recognition of cancer-related lesions with clinical priors on white-light endoscopy

Deep learning-based computer-aided diagnosis techniques have demonstrated encouraging performance in endoscopic lesion identification and detection, and have reduced the rate of missed and false detections of disease during endoscopy. However, the interpretability of the model-based results has not been adequately addressed by existing methods. This phenomenon is directly manifested by a significant bias in the representation of feature localization. Good recognition models experience severe feature localization errors, particularly for lesions with subtle morphological features, and such unsatisfactory performance hinders the clinical deployment of models. To effectively alleviate this problem, we proposed a solution to optimize the localization bias in feature representations of cancer-related recognition models that is difficult to accurately label and identify in clinical practice. Optimization was performed in the training phase of the model through the proposed data augmentation method and auxiliary loss function based on clinical priors. The data augmentation method, called partial jigsaw, can "break" the spatial structure of lesion-independent image blocks and enrich the data feature space to decouple the interference of background features on the space and focus on fine-grained lesion features. The annotation-based auxiliary loss function used class activation maps for sample distribution correction and led the model to present localization representation converging on the gold standard annotation of visualization maps. The results show that with the improvement of our method, the precision of model recognition reached an average of 92.79%, an F1-score of 92.61%, and accuracy of 95.56% based on a dataset constructed from 23 hospitals. In addition, we quantified the evaluation representation of visualization feature maps. The improved model yielded significant offset correction results for visualized feature maps compared with the baseline model. The average visualization-weighted positive coverage improved from 51.85% to 83.76%. The proposed approach did not change the deployment capability and inference speed of the original model and can be incorporated into any state-of-the-art neural network. It also shows the potential to provide more accurate localization inference results and assist in clinical examinations during endoscopies.

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.

Exploring and Characterizing Patient Multibehavior Engagement Trails and Patient Behavior Preference Patterns in Pathway-Based mHealth Hypertension Self-Management: Analysis of Use Data

Background

Hypertension is a long-term medical condition. Mobile health (mHealth) services can help out-of-hospital patients to self-manage. However, not all management is effective, possibly because the behavior mechanism and behavior preferences of patients with various characteristics in hypertension management were unclear.

Objective

The purpose of this study was to (1) explore patient multibehavior engagement trails in the pathway-based hypertension self-management, (2) discover patient behavior preference patterns, and (3) identify the characteristics of patients with different behavior preferences.

Methods

This study included 863 hypertensive patients who generated 295,855 use records in the mHealth app from December 28, 2016, to July 2, 2020. Markov chain was used to infer the patient multibehavior engagement trails, which contained the type, quantity, time spent, sequence, and transition probability value (TP value) of patient behavior. K-means algorithm was used to group patients by the normalized behavior preference features: the number of behavioral states that a patient performed in each trail. The pages in the app represented the behavior states. Chi-square tests, Z-test, analyses of variance, and Bonferroni multiple comparisons were conducted to characterize the patient behavior preference patterns.

Results

Markov chain analysis revealed 3 types of behavior transition (1-way transition, cycle transition, and self-transition) and 4 trails of patient multibehavior engagement. In perform task trail (PT-T), patients preferred to start self-management from the states of task blood pressure (BP), task drug, and task weight (TP value 0.29, 0.18, and 0.20, respectively), and spent more time on the task food state (35.87 s). Some patients entered the states of task BP and task drug (TP value 0.20, 0.25) from the reminder item state. In the result-oriented trail (RO-T), patients spent more energy on the ranking state (19.66 s) compared to the health report state (13.25 s). In the knowledge learning trail (KL-T), there was a high probability of cycle transition (TP value 0.47, 0.31) between the states of knowledge list and knowledge content. In the support acquisition trail (SA-T), there was a high probability of self-transition in the questionnaire (TP value 0.29) state. Cluster analysis discovered 3 patient behavior preference patterns: PT-T cluster, PT-T and KL-T cluster, and PT-T and SA-T cluster. There were statistically significant associations between the behavior preference pattern and gender, education level, and BP.

Conclusions

This study identified the dynamic, longitudinal, and multidimensional characteristics of patient behavior. Patients preferred to focus on BP, medications, and weight conditions and paid attention to BP and medications using reminders. The diet management and questionnaires were complicated and difficult to implement and record. Competitive methods such as ranking were more likely to attract patients to pay attention to their own self-management states. Female patients with lower education level and poorly controlled BP were more likely to be highly involved in hypertension health education.

Prediction of central venous catheter-associated deep venous thrombosis in pediatric critical care settings

BACKGROUND

An increase in the incidence of central venous catheter (CVC)-associated deep venous thrombosis (CADVT) has been reported in pediatric patients over the past decade. At the same time, current screening guidelines for venous thromboembolism risk have low sensitivity for CADVT in hospitalized children. This study utilized a multimodal deep learning model to predict CADVT before it occurs.

METHODS

Children who were admitted to intensive care units (ICUs) between December 2015 and December 2018 and with CVC placement at least 3 days were included. The variables analyzed included demographic characteristics, clinical conditions, laboratory test results, vital signs and medications. A multimodal deep learning (MMDL) model that can handle temporal data using long short-term memory (LSTM) and gated recurrent units (GRUs) was proposed for this prediction task. Four benchmark machine learning models, logistic regression (LR), random forest (RF), gradient boosting decision tree (GBDT) and a published cutting edge MMDL, were used to compare and evaluate the models with a fivefold cross-validation approach. Accuracy, recall, area under the ROC curve (AUC), and average precision (AP) were used to evaluate the discrimination of each model at three time points (24 h, 48 h and 72 h) before CADVT occurred. Brier score and Spiegelhalter's z test were used measure the calibration of these prediction models.

RESULTS

A total of 1830 patients were included in this study, and approximately 15% developed CADVT. In the CADVT prediction task, the model proposed in this paper significantly outperforms both traditional machine learning models and existing multimodal deep learning models at all 3 time points. It achieved 77% accuracy and 90% recall at 24 h before CADVT was discovered. It can be used to accurately predict the occurrence of CADVT 72 h in advance with an accuracy of greater than 75%, a recall of more than 87%, and an AUC value of 0.82.

CONCLUSION

In this study, a machine learning method was successfully established to predict CADVT in advance. These findings demonstrate that artificial intelligence (AI) could provide measures for thromboprophylaxis in a pediatric intensive care setting.

A home-based pulmonary rehabilitation mHealth system to enhance the exercise capacity of patients with COPD: development and evaluation

BACKGROUND

Patients with chronic obstructive pulmonary disease (COPD) experience deficits in exercise capacity and physical activity as their disease progresses. Pulmonary rehabilitation (PR) can enhance exercise capacity of patients and it is crucial for patients to maintain a lifestyle which is long-term physically active. This study aimed to develop a home-based rehabilitation mHealth system incorporating behavior change techniques (BCTs) for COPD patients, and evaluate its technology acceptance and feasibility.

METHODS

Guided by the medical research council (MRC) framework the process of this study was divided into four steps. In the first step, the prescription was constructed. The second step was to formulate specific intervention functions based on the behavior change wheel theory. Subsequently, in the third step we conducted iterative system development. And in the last step two pilot studies were performed, the first was for the improvement of system functions and the second was to explore potential clinical benefits and validate the acceptance and usability of the system.

RESULTS

A total of 17 participants were enrolled, among them 12 COPD participants completed the 12-week study. For the clinical outcomes, Six-Minute Walk Test (6MWT) showed significant difference (P = .023) over time with an improvement exceeded the minimal clinically important difference (MCID). Change in respiratory symptom (CAT score) was statistically different (P = .031) with a greater decrease of - 3. The mMRC levels reduced overall and showed significant difference. The overall compliance of this study reached 82.20% (± 1.68%). The results of questionnaire and interviews indicated good technology acceptance and functional usability. The participants were satisfied with the mHealth-based intervention.

CONCLUSIONS

This study developed a home-based PR mHealth system for COPD patients. We showed that the home-based PR mHealth system incorporating BCTs is a feasible and acceptable intervention for COPD patients, and COPD patients can benefit from the intervention delivered by the system. The proposed system played an important auxiliary role in offering exercise prescription according to the characteristics of patients. It provided means and tools for further individuation of exercise prescription in the future.

Building an information system to facilitate pharmacogenomics clinical translation with clinical decision support

Pharmacogenomics clinical decision support (PGx-CDS) is an important tool to incorporate PGx information into existing clinical workflows and facilitate PGx clinical translation. However, due to the lack of a computable formalization to represent the primary PGx knowledge, the complexity of genomics information and the lag of current commercial electronic health record (EHR) system for precision medicine, it is difficult to develop computerized PGx-CDS. Therefore, we explored a novel approach to build an information system, named the Pharmacogenomics Clinical Translation Platform (PCTP), for PGx clinical implementation. The PCTP can represent, store, and manage the primary PGx knowledge in a structured and computable format. Moreover, it has the potential to provide various PGx-CDS services and simplify the integration of PGx-CDS into EHRs.

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.

A Visual Phenotype-Based Differential Diagnosis Process for Rare Diseases

PURPOSE

Phenotype-based rapid diagnosis can make up for the time-consuming genetic sequencing diagnosis of rare diseases. However, the collected phenotypes of patients can sometimes be inaccurate or incomplete, which limits the accuracy of diagnostic results. To solve this problem, we try to design a phenotype-based differential diagnosis process for rare diseases to achieve rapid and accurate diagnosis of rare diseases.

METHODS

The core of the differential diagnosis of rare diseases is to optimize the phenotype information of a specific patient and the visualized comparative analysis of diseases. To recommend additional phenotypes, replace the fuzzy phenotypes and filter the unexplained phenotypes for patients, we constructed a phenotype hierarchical network and a disease-phenotype differential network and calculated the phenotype co-occurrence relationship. In addition, we designed a visual comparative analysis method to explore the correlation and difference of disease phenotypes.

RESULTS

The evaluation based on the published 10 rare disease cases demonstrated that after the optimization of patient phenotype information through our differential diagnosis, the target disease often got a better ranking and recommendation score than before. We have deployed this scheme on the RDmap project ( http://rdmap.nbscn.org ).

CONCLUSION

Compared to genetic and molecular analysis, phenotype-based diagnosis is faster, cheaper, and easier. The differential diagnosis process we designed can optimize the phenotype information of patients and better locate the target disease. It can also help to make screening decisions before genetic testing.

A Patient-Screening Tool for Clinical Research Based on Electronic Health Records Using OpenEHR: Development Study

Background

The widespread adoption of electronic health records (EHRs) has facilitated the secondary use of EHR data for clinical research. However, screening eligible patients from EHRs is a challenging task. The concepts in eligibility criteria are not completely matched with EHRs, especially derived concepts. The lack of high-level expression of Structured Query Language (SQL) makes it difficult and time consuming to express them. The openEHR Expression Language (EL) as a domain-specific language based on clinical information models shows promise to represent complex eligibility criteria.

Objective

The study aims to develop a patient-screening tool based on EHRs for clinical research using openEHR to solve concept mismatch and improve query performance.

Methods

A patient-screening tool based on EHRs using openEHR was proposed. It uses the advantages of information models and EL in openEHR to provide high-level expressions and improve query performance. First, openEHR archetypes and templates were chosen to define concepts called simple concepts directly from EHRs. Second, openEHR EL was used to generate derived concepts by combining simple concepts and constraints. Third, a hierarchical index corresponding to archetypes in Elasticsearch (ES) was generated to improve query performance for subqueries and join queries related to the derived concepts. Finally, we realized a patient-screening tool for clinical research.

Results

In total, 500 sentences randomly selected from 4691 eligibility criteria in 389 clinical trials on stroke from the Chinese Clinical Trial Registry (ChiCTR) were evaluated. An openEHR-based clinical data repository (CDR) in a grade A tertiary hospital in China was considered as an experimental environment. Based on these, 589 medical concepts were found in the 500 sentences. Of them, 513 (87.1%) concepts could be represented, while the others could not be, because of a lack of information models and coarse-grained requirements. In addition, our case study on 6 queries demonstrated that our tool shows better query performance among 4 cases (66.67%).

Conclusions

We developed a patient-screening tool using openEHR. It not only helps solve concept mismatch but also improves query performance to reduce the burden on researchers. In addition, we demonstrated a promising solution for secondary use of EHR data using openEHR, which can be referenced by other researchers.

Patterns for Patient Engagement with the Hypertension Management and Effects of Electronic Health Care Provider Follow-up on These Patterns: Cluster Analysis

Background

Hypertension is a long-term medical condition. Electronic and mobile health care services can help patients to self-manage this condition. However, not all management is effective, possibly due to different levels of patient engagement (PE) with health care services. Health care provider follow-up is an intervention to promote PE and blood pressure (BP) control.

Objective

This study aimed to discover and characterize patterns of PE with a hypertension self-management app, investigate the effects of health care provider follow-up on PE, and identify the follow-up effects on BP in each PE pattern.

Methods

PE was represented as the number of days that a patient recorded self-measured BP per week. The study period was the first 4 weeks for a patient to engage in the hypertension management service. K-means algorithm was used to group patients by PE. There was compliance follow-up, regular follow-up, and abnormal follow-up in management. The follow-up effect was calculated by the change in PE (CPE) and the change in systolic blood pressure (CSBP, SBP) before and after each follow-up. Chi-square tests and z scores were used to ascertain the distribution of gender, age, education level, SBP, and the number of follow-ups in each cluster. The follow-up effect was identified by analysis of variances. Once a significant effect was detected, Bonferroni multiple comparisons were further conducted to identify the difference between 2 clusters.

Results

Patients were grouped into 4 clusters according to PE: (1) PE started low and dropped even lower (PELL), (2) PE started high and remained high (PEHH), (3) PE started high and dropped to low (PEHL), and (4) PE started low and rose to high (PELH). Significantly more patients over 60 years old were found in the PEHH cluster (P≤.05). Abnormal follow-up was significantly less frequent (P≤.05) in the PELL cluster. Compliance follow-up and regular follow-up can improve PE. In the clusters of PEHH and PELH, the improvement in PE in the first 3 weeks and the decrease in SBP in all 4 weeks were significant after follow-up. The SBP of the clusters of PELL and PELH decreased more (–6.1 mmHg and –8.4 mmHg) after follow-up in the first week.

Conclusions

Four distinct PE patterns were identified for patients engaging in the hypertension self-management app. Patients aged over 60 years had higher PE in terms of recording self-measured BP using the app. Once SBP reduced, patients with low PE tended to stop using the app, and a continued decline in PE occurred simultaneously with the increase in SBP. The duration and depth of the effect of health care provider follow-up were more significant in patients with high or increased engagement after follow-up.

Explainable machine-learning predictions for complications after pediatric congenital heart surgery

The quality of treatment and prognosis after pediatric congenital heart surgery remains unsatisfactory. A reliable prediction model for postoperative complications of congenital heart surgery patients is essential to enable prompt initiation of therapy and improve the quality of prognosis. Here, we develop an interpretable machine-learning-based model that integrates patient demographics, surgery-specific features and intraoperative blood pressure data for accurately predicting complications after pediatric congenital heart surgery. We used blood pressure variability and the k-means algorithm combined with a smoothed formulation of dynamic time wrapping to extract features from time-series data. In addition, SHAP framework was used to provide explanations of the prediction. Our model achieved the best performance both in binary and multi-label classification compared with other consensus-based risk models. In addition, this explainable model explains why a prediction was made to help improve the clinical understanding of complication risk and generate actionable knowledge in practice. The combination of model performance and interpretability is easy for clinicians to trust and provide insight into how they should respond before the condition worsens after pediatric congenital heart surgery.

Simultaneous Recognition of Atrophic Gastritis and Intestinal Metaplasia on White Light Endoscopic Images Based on Convolutional Neural Networks: A Multicenter Study

INTRODUCTION

Patients with atrophic gastritis (AG) or gastric intestinal metaplasia (GIM) have elevated risk of gastric adenocarcinoma. Endoscopic screening and surveillance have been implemented in high incidence countries. The study aimed to evaluate the accuracy of a deep convolutional neural network (CNN) for simultaneous recognition of AG and GIM.

METHODS

Archived endoscopic white light images with corresponding gastric biopsies were collected from 14 hospitals located in different regions of China. Corresponding images by anatomic sites containing AG, GIM, and chronic non-AG were categorized using pathology reports. The participants were randomly assigned (8:1:1) to the training cohort for developing the CNN model (TResNet), the validation cohort for fine-tuning, and the test cohort for evaluating the diagnostic accuracy. The area under the curve (AUC), sensitivity, specificity, and accuracy with 95% confidence interval (CI) were calculated.

RESULTS

A total of 7,037 endoscopic images from 2,741 participants were used to develop the CNN for recognition of AG and/or GIM. The AUC for recognizing AG was 0.98 (95% CI 0.97-0.99) with sensitivity, specificity, and accuracy of 96.2% (95% CI 94.2%-97.6%), 96.4% (95% CI 94.8%-97.9%), and 96.4% (95% CI 94.4%-97.8%), respectively. The AUC for recognizing GIM was 0.99 (95% CI 0.98-1.00) with sensitivity, specificity, and accuracy of 97.9% (95% CI 96.2%-98.9%), 97.5% (95% CI 95.8%-98.6%), and 97.6% (95% CI 95.8%-98.6%), respectively.

DISCUSSION

CNN using endoscopic white light images achieved high diagnostic accuracy in recognizing AG and GIM.

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.

Prediction of complications after paediatric cardiac surgery

OBJECTIVES

Our objectives were to identify the risk factors for postoperative complications after paediatric cardiac surgery, develop a tool for predicting postoperative complications and compare it with other risk adjustment tools of congenital heart disease.

METHODS

A total of 2308 paediatric patients who had undergone cardiac surgeries with cardiopulmonary bypass support in a single centre were included in this study. A univariate analysis was performed to determine the association between perioperative variables and postoperative complications. Statistically significant variables were integrated into a synthetic minority oversampling technique-based XGBoost model which is an implementation of gradient boosted decision trees designed for speed and performance. The 7 traditional risk assessment tools used to generate the logistic regression model as the benchmark in the evaluation included the Aristotle Basic score and category, Risk Adjustment for Congenital Heart Surgery (RACHS-1), Society of Thoracic Surgeons-European Association for Cardio-Thoracic Surgery (STS-EACTS) mortality score and category and STS morbidity score and category.

RESULTS

Our XGBoost prediction model showed the best prediction performance (area under the receiver operating characteristic curve = 0.82) when compared with these risk adjustment models. However, all of these models exhibited a relatively lower sensitivity due to imbalanced classes. The sensitivity of our optimization approach (synthetic minority oversampling technique-based XGBoost) was 0.74, which was significantly higher than the average sensitivity of the traditional models of 0.26. Furthermore, the postoperative length of hospital stay, length of cardiac intensive care unit stay and length of mechanical ventilation duration were significantly increased for patients who experienced postoperative complications.

CONCLUSIONS

Postoperative complications of paediatric cardiac surgery can be predicted based on perioperative data using our synthetic minority oversampling technique-based XGBoost model before deleterious outcomes ensue.

Combining gene expression signature with clinical features for survival stratification of gastric cancer

The AJCC staging system is considered as the golden standard in clinical practice. However, it remains some pitfalls in assessing the prognosis of gastric cancer (GC) patients with similar clinicopathological characteristics. We aim to develop a new clinic and genetic risk score (CGRS) to improve the prognosis prediction of GC patients. We established genetic risk score (GRS) based on nine-gene signature including APOD, CCDC92, CYS1, GSDME, ST8SIA5, STARD3NL, TIMEM245, TSPYL5, and VAT1 based on the gene expression profiles of the training set from the Asian Cancer Research Group (ACRG) cohort by LASSO-Cox regression algorithms. CGRS was established by integrating GRS with clinical risk score (CRS) derived from Surveillance, Epidemiology, and End Results (SEER) database. GRS and CGRS dichotomized GC patients into high and low risk groups with significantly different prognosis in four independent cohorts with different data types, such as microarray, RNA sequencing and qRT-PCR (all HR > 1, all P < 0.001). Both GRS and CGRS were prognostic signatures independent of the AJCC staging system. Receiver operating characteristic (ROC) analysis showed that area under ROC curve of CGRS was larger than that of the AJCC staging system in most cohorts we studied. Nomogram and web tool (http://39.100.117.92/CGRS/) based on CGRS were developed for clinicians to conveniently assess GC prognosis in clinical practice. CGRS integrating genetic signature with clinical features shows strong robustness in predicting GC prognosis, and can be easily applied in clinical practice through the web application.