Attention-assisted hybrid CNN-BILSTM-BiGRU model with SMOTE-Tomek method to detect cardiac arrhythmia based on 12-lead electrocardiogram signals.

Objectives: Cardiac arrhythmia is one of the most severe cardiovascular diseases that can be fatal. Therefore, its early detection is critical. However, detecting types of arrhythmia by physicians based on visual identification is time-consuming and subjective. Deep learning can develop effective approaches to classify arrhythmias accurately and quickly. This study proposed a deep learning approach developed based on a Chapman-Shaoxing electrocardiogram (ECG) dataset signal to detect seven types of arrhythmias. Method: Our DNN model is a hybrid CNN-BILSTM-BiGRU algorithm assisted by a multi-head self-attention mechanism regarding the challenging problem of classifying various arrhythmias of ECG signals. Additionally, the synthetic minority oversampling technique (SMOTE)-Tomek technique was utilized to address the data imbalance problem to detect and classify cardiac arrhythmias. Result: The proposed model, trained with a single lead, was tested using a dataset containing 10,466 participants. The performance of the algorithm was evaluated using a random split validation approach. The proposed algorithm achieved an accuracy of 98.57% by lead II and 98.34% by lead aVF for the classification of arrhythmias. Conclusion: We conducted an analysis of single-lead ECG signals to evaluate the effectiveness of our proposed hybrid model in diagnosing and classifying different types of arrhythmias. We trained separate classification models using each individual signal lead. Additionally, we implemented the SMOTE-Tomek technique along with cross-entropy loss as a cost function to address the class imbalance problem. Furthermore, we utilized a multi-headed self-attention mechanism to adjust the network structure and classify the seven arrhythmia classes. Our model achieved high accuracy and demonstrated good generalization ability in detecting ECG arrhythmias. However, further testing of the model with diverse datasets is crucial to validate its performance.

Non-invasive fractional flow reserve estimation using deep learning on intermediate left anterior descending coronary artery lesion angiography images.

This study aimed to design an end-to-end deep learning model for estimating the value of fractional flow reserve (FFR) using angiography images to classify left anterior descending (LAD) branch angiography images with average stenosis between 50 and 70% into two categories: FFR > 80 and FFR ≤ 80. In this study 3625 images were extracted from 41 patients' angiography films. Nine pre-trained convolutional neural networks (CNN), including DenseNet121, InceptionResNetV2, VGG16, VGG19, ResNet50V2, Xception, MobileNetV3Large, DenseNet201, and DenseNet169, were used to extract the features of images. DenseNet169 indicated higher performance compared to other networks. AUC, Accuracy, Sensitivity, Specificity, Precision, and F1-score of the proposed DenseNet169 network were 0.81, 0.81, 0.86, 0.75, 0.82, and 0.84, respectively. The deep learning-based method proposed in this study can non-invasively and consistently estimate FFR from angiographic images, offering significant clinical potential for diagnosing and treating coronary artery disease by combining anatomical and physiological parameters.

Features and functions of decision support systems for appropriate diagnostic imaging: a scoping review.

BACKGROUND: Diagnostic imaging decision support (DI-DS) systems could be effective tools for reducing inappropriate diagnostic imaging examinations. Since effective design and evaluation of these systems requires in-depth understanding of their features and functions, the present study aims to map the existing literature on DI-DS systems to identify features and functions of these systems. METHODS: The search was performed using Scopus, Embase, PubMed, Web of Science, and Cochrane Central Registry of Controlled Trials (CENTRAL) and was limited to 2000 to 2021. Analytical studies, descriptive studies, reviews and book chapters that explicitly addressed the functions or features of DI-DS systems were included. RESULTS: A total of 6,046 studies were identified. Out of these, 55 studies met the inclusion criteria. From these, 22 functions and 22 features were identified. Some of the identified features were: visibility, content chunking/grouping, deployed as a multidisciplinary program, clinically valid and relevant feedback, embedding current evidence, and targeted recommendations. And, some of the identified functions were: displaying an appropriateness score, recommending alternative or more appropriate imaging examination(s), providing recommendations for next diagnostic steps, and providing safety alerts. CONCLUSIONS: The set of features and functions obtained in the present study can provide a basis for developing well-designed DI-DS systems, which could help to improve adherence to diagnostic imaging guidelines, minimize unnecessary costs, and improve the outcome of care through appropriate diagnosis and on-time care delivery.

Effects of Guideline-based Computerized Provider Order Entry Systems on the Chemotherapy Order Process: a Systematic Review.

Background: Computerized Provider Order Entry (CPOE) systems developed based on clinical guidelines are believed to greatly reduce chemotherapy medication prescription errors. Objective: The present study reviewed the effects of guideline-based CPOEs on the chemotherapy order process. Methods: PubMed, Scopus, Embase, Web of Science, and IEEE Xplore databases published up to 1 June 2020 were systematically searched for studies investigating the effect of guideline-based CPOEs on the chemotherapy order process. Moreover, the bibliography of relevant retrieved publications was also checked. Results: Nineteen articles from the five databases met the eligibility criteria and were reviewed. Eleven out of 19 (58%) articles investigated the effect of CPOEs on medication errors, and other studies examined other aspects of CPOE efficacy, including time required for chemotherapy prescriptions; Safety, policy compliance and communication between health care providers; physicians prescribing behavior; quality and safety of treatment; workflow; direct patient care time; and adherence to guidelines. In addition, 15 out of 19 mentioned the use of specific clinical guidelines. Conclusion: Evidence indicates CPOEs can positively affect the quality of healthcare service delivery for cancer patients, but there is still a dearth of clinical outcome evaluation data about the effects of these systems on patients undergoing chemotherapy. Moreover, there is limited information about guideline compliance errors, which highlights the needs for further research in this area.

Development of an online cancer data collection and processing tool for population-based cancer registries in a low-resource setting: The CanDCap experience from Golestan, Iran.

BACKGROUND: Golestan Population-based Cancer Registry (GPCR) with more than 15-years experiences developed an in-house online software called Cancer Data Collection and Processing (CanDCap) to improve its data collection operations from the conventional offline method to new online method. We aimed to report the methods and framework that GPCR applied to design and implementation of the CanDCap. METHODS: CanDCap was designed based on International Agency for Research on Cancer (IARC) protocols and standards and according to the GPCR workflow. CanDCap has two parts including a web-based online part for data collection and a windows-based part for data processing consisting of quality control and deduplication of repeated records. Questionnaire for User Interface Satisfaction (QUIS) was used in order to assess user interaction satisfaction. RESULTS: CanDCap was implemented in 2018 and could improve the quality of the GPCR data during its first three years of activity (2018-2020), during which about 9,000 records were registered. The coverage for optional items including national ID, father name, address and telephone number were improved from 23 %, 32 %, 83 % and 82 % in conventional offline method (2015-2017) to 83 %, 81 %, 87 %, and 90 % after using the CanDCap (2018-2020), respectively. The timeliness was also improved from 4 years to 2 years. Overall, user interaction satisfaction was acceptable (7.8 out of 9). CONCLUSION: CanDCap could resulted in improvement in data quality and timeliness of the GPCR as a cancer registry unit with limited resources. It has the potential to be considered as a model for population-based cancer registries in lower-resource settings.

Developing a Prognostic Model to Predict Mortality in Patients with Acute Bacterial Meningitis.

Bacterial meningitis is one of the harmful and deadly infectious diseases, and any delay in its treatment will lead to death. In this paper, a prognostic model was developed to predict the risk of death amongst probable cases of bacterial meningitis. Our prognostic model was developed using a decision tree algorithm on the national meningitis registry of the Iranian Center for Disease and Prevention (ICDCP) containing 3,923 records of meningitis suspected cases in 2018-2019. The most important features have been selected for the model construction. This model can predict the mortality risk for the meningitis probable cases with 78% accuracy, 84% sensitivity, and 73% specificity. The identified variables in prognosis the death included age and CSF protein level. CSF protein level (mg/dl) <= 65 versus > 65 provided the first branch of our decision tree. The highest mortality risk (85.8%) was seen in the patients >65 CSF protein level with 30 years < of age. For the patients <=30 year of age with CSF protein level >137 (mg/dl), the mortality risk was 60%. The prognostic factors identified in the present study draw the attention of clinicians to provide early specific measures, such as the admission of patients with a higher risk of death to intensive care units (ICU). It could also provide a helpful risk score tool in decision-making in the early phases of admission in pandemics, decrease mortality rate and improve public health operations efficiently in infectious diseases.

An Intelligent Clinical Decision Support System for Predicting Acute Graft-versus-host Disease (aGvHD) following Allogeneic Hematopoietic Stem Cell Transplantation.

BACKGROUND: Acute graft-versus-host disease (aGvHD) is a complex and often multisystem disease that causes morbidity and mortality in 35% of patients receiving allogeneic hematopoietic stem cell transplantation (AHSCT). OBJECTIVE: This study aimed to implement a Clinical Decision Support System (CDSS) for predicting aGvHD following AHSCT on the transplantation day. MATERIAL AND METHODS: In this developmental study, the data of 182 patients with 31 attributes, which referred to Taleghani Hospital Tehran, Iran during 2009-2017, were analyzed by machine learning (ML) algorithms which included XGBClassifier, HistGradientBoostingClassifier, AdaBoostClassifier, and RandomForestClassifier. The criteria measurement used to evaluate these algorithms included accuracy, sensitivity, and specificity. Using the machine learning developed model, a CDSS was implemented. The performance of the CDSS was evaluated by Cohen's Kappa coefficient. RESULTS: Of the 31 included variables, albumin, uric acid, C-reactive protein, donor age, platelet, lactate Dehydrogenase, and Hemoglobin were identified as the most important predictors. The two algorithms XGBClassifier and HistGradientBoostingClassifier with an average accuracy of 90.70%, sensitivity of 92.5%, and specificity of 89.13% were selected as the most appropriate ML models for predicting aGvHD. The agreement between CDSS prediction and patient outcome was 92%. CONCLUSION: ML methods can reliably predict the likelihood of aGvHD at the time of transplantation. These methods can help us to limit the number of risk factors to those that have significant effects on the outcome. However, their performance is heavily dependent on selecting the appropriate methods and algorithms. The next generations of CDSS may use more and more machine learning approaches.

Machine Learning Classification Algorithms to Predict aGvHD following Allo-HSCT: A Systematic Review.

BACKGROUND: The acute graft-versus-host disease (aGvHD) is the most important cause of mortality in patients receiving allogeneic hematopoietic stem cell transplantation. Given that it occurs at the stage of severe tissue damage, its diagnosis is late. With the advancement of machine learning (ML), promising real-time models to predict aGvHD have emerged. OBJECTIVE: This article aims to synthesize the literature on ML classification algorithms for predicting aGvHD, highlighting algorithms and important predictor variables used. METHODS: A systemic review of ML classification algorithms used to predict aGvHD was performed using a search of the PubMed, Embase, Web of Science, Scopus, Springer, and IEEE Xplore databases undertaken up to April 2019 based on Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) statements. The studies with a focus on using the ML classification algorithms in the process of predicting of aGvHD were considered. RESULTS: After applying the inclusion and exclusion criteria, 14 studies were selected for evaluation. The results of the current analysis showed that the algorithms used were Artificial Neural Network (79%), Support Vector Machine (50%), Naive Bayes (43%), k-Nearest Neighbors (29%), Regression (29%), and Decision Trees (14%), respectively. Also, many predictor variables have been used in these studies so that we have divided them into more abstract categories, including biomarkers, demographics, infections, clinical, genes, transplants, drugs, and other variables. CONCLUSION: Each of these ML algorithms has a particular characteristic and different proposed predictors. Therefore, it seems these ML algorithms have a high potential for predicting aGvHD if the process of modeling is performed correctly.