scholarly journals Accurate Severe vs Non-severe COVID-19 Clinical Type Classification: a Multimodality Machine Learning Study (Preprint)

2020 ◽  
Author(s):  
Yuanfang Chen ◽  
Liu Ouyang ◽  
Forrest S. Bao ◽  
Qian Li ◽  
Lei Han ◽  
...  
Keyword(s):  
Machine Learning ◽  
Laboratory Testing ◽  
Predictive Accuracy ◽  
Clinical Information ◽  
Clinical Decision ◽  
Immunological Responses ◽  
Self Evaluation ◽  
Optimal Outcomes ◽  
Clinical Type ◽  
Multimodal Features

BACKGROUND Effectively and efficiently diagnosing COVID-19 patients with accurate clinical type is essential to achieve optimal outcomes for the patients as well as reducing the risk of overloading the healthcare system. Currently, severe and non-severe COVID-19 types are differentiated by only a few features, which do not comprehensively characterize the complicated pathological, physiological, and immunological responses to SARS-CoV-2 invasion in different types. In addition, these type-defining features may not be readily testable at time of diagnosis. OBJECTIVE This study aimed to accurately differentiate severe and non-severe COVID-19 clinical types based on multiple medical features and provide reliable predictions for clinical decision support. METHODS In this study, we recruited 214 confirmed COVID-19 patients in non-severe and 148 in severe type. The patients’ clinical (including 26 features), and laboratory testing results (26 features) upon admission were acquired as two input modalities. Exploratory analyses demonstrated that these features differed substantially between two clinical types. Machine learning random forest (RF) models based on all features in each modality as well as top 5 features in each modality combined were developed and validated to differentiate COVID-19 clinical types. RESULTS Using clinical and laboratory results as input independently, RF models achieved 90% and 95% predictive accuracy, respectively. Input features’ importance scores were further evaluated and top five features from each modality were identified (age, hypertension, cardiovascular disease, gender, diabetes; D-Dimer, hsTNI, absolute neutrophil count, IL-6, and LDH, in descending order). Using these top 10 multimodal features as the only input instead of all 52 features combined, RF model was able to achieve 99% predictive accuracy. CONCLUSIONS These findings shed light on how the human body reacts to SARS-CoV-2 invasion as a unity and provide insights on effectively evaluating COVID-19 patient’s severity based on more common medical features when gold-standard features were not available. We suggest that clinical information can be used as an initial screening tool for self-evaluation and triaging, while laboratory testing results are applied when accuracy is the priority.

scholarly journals An Interpretable Machine Learning Framework for Accurate Severe vs Non-severe COVID-19 Clinical Type Classification

2020 ◽  
Author(s):  
Yuanfang Chen ◽  
Liu Ouyang ◽  
Sheng Bao ◽  
Qian Li ◽  
Lei Han ◽  
...  
Keyword(s):  
Machine Learning ◽  
Predictive Accuracy ◽  
Immunological Responses ◽  
Optimal Outcomes ◽  
Learning Framework ◽  
Clinical Type ◽  
Multimodal Features ◽  
Treatment Plans ◽  
Severe Type ◽  
Type Classification

Effectively and efficiently diagnosing COVID-19 patients with accurate clinical type is essential to achieve optimal outcomes of the patients as well as reducing the risk of overloading the healthcare system. Currently, severe and non-severe COVID-19 types are differentiated by only a few clinical features, which do not comprehensively characterize complicated pathological, physiological, and immunological responses to SARS-CoV-2 invasion in different types. In this study, we recruited 214 confirmed COVID-19 patients in non-severe and 148 in severe type, from Wuhan, China. The patients' comorbidity and symptoms (26 features), and blood biochemistry (26 features) upon admission were acquired as two input modalities. Exploratory analyses demonstrated that these features differed substantially between two clinical types. Machine learning random forest (RF) models using features in each modality were developed and validated to classify COVID-19 clinical types. Using comorbidity/symptom and biochemistry as input independently, RF models achieved >90% and >95% predictive accuracy, respectively. Input features' importance based on Gini impurity were further evaluated and top five features from each modality were identified (age, hypertension, cardiovascular disease, gender, diabetes; D-Dimer, hsTNI, neutrophil, IL-6, and LDH). Combining top 10 multimodal features, RF model achieved >99% predictive accuracy. These findings shed light on how the human body reacts to SARS-CoV-2 invasion as a unity and provide insights on effectively evaluating COVID-19 patient's severity and developing treatment plans accordingly. We suggest that symptoms and comorbidities can be used as an initial screening tool for triaging, while biochemistry and features combined are applied when accuracy is the priority.

scholarly journals 409 Dynamic Phenotype Learning: A Novel Machine Learning Approach To Develop And Discover New OSA Sub-Types

SLEEP ◽  
2021 ◽  
Vol 44 (Supplement_2) ◽  
pp. A162-A163
Author(s):  
Chris Fernandez ◽  
Sam Rusk ◽  
Yoav Nygate ◽  
Nick Glattard ◽  
Jessica Arguelles ◽  
...  
Keyword(s):  
Machine Learning ◽  
Deep Learning ◽  
Predictive Accuracy ◽  
Care Pathway ◽  
Clinical Decision ◽  
Current Approach ◽  
Sleep Study ◽  
Personalized Care ◽  
Obstructive Sleep ◽  
Patient Reported

Abstract Introduction Current approach to processing polysomnography is labor intensive and produces metrics that are poor at identifying obstructive sleep apnea (OSA) phenotypes necessary to enhance personalized care. We describe our approach to utilize Dynamic Phenotype Learning (DPL) as an innovative machine learning technique to identify OSA subtypes that can better predict clinical risk and success with therapies. Methods This study is a collaboration between Kaiser Permanente Southern California (KPSC), a large integrated health system, and EnsoData Research, which specializes in applied A.I. analysis of physiologic waveforms. KPSC sleep medicine compiled a database of N=5,368–234,250 subjects that include Types I, II, III, or IV sleep study data, daily PAP data, patient reported data, and comprehensive electronic health record information, with present research applications to study the relationship between OSA and PAP adherence with cardiovascular outcomes, health economic impacts, novel coronavirus (COVID-19) outcomes, and predictive PAP adherence and OSA severity clinical decision tools. DPL is a machine learning method for studying known and new biomarkers and care-pathway indices, including personalized screening, diagnostic, treatment, adherence, and outcomes predictors, that can be rooted in physiologic data. DPL processes waveform signal data without scoring, annotations, or expert synthesis, by applying a novel machine learning mechanism that blurs supervised and unsupervised deep learning paradigms, to find relationships between physiome dynamics expressed in waveforms and phenotypes and endotypes of interest. Results We demonstrate DPL method with an illustrative study on known indices, to explain its ability to (1) lift theoretical-empirical predictive accuracy ceilings and (b) reduce several sources of bias and variance. We show DPL exceeds the ROC-AUC and PRC-AUC of equivalent deep learning models in N=30,000 Report-Demographic (ODI, PLMSI, Weight), Scoring (REM, OSA), and Waveform (EEG, PPG) datasets respectively to predict AHI, TST, Brain Age, and OSA-Insomnia. We present our current collaboration advancing DPL to identify specific phenotypes that better predict: (a) cardiovascular risk; (b) neurocognitive outcomes; (c) response to PAP and alternative therapies. Conclusion DPL methods are being applied to large and comprehensive patient dataset to identify phenotypic indices and biomarkers with potential to take us beyond the AHI, and uncover relationships between OSA sub-types, treatments, and health outcomes. Support (if any):

scholarly journals Identifying signals associated with psychiatric illness utilizing language and images posted to Facebook

2020 ◽  
Vol 6 (1) ◽  
Author(s):  
Michael L. Birnbaum ◽  
Raquel Norel ◽  
Anna Van Meter ◽  
Asra F. Ali ◽  
Elizabeth Arenare ◽  
...  
Keyword(s):  
Machine Learning ◽  
Clinical Decision Making ◽  
Negative Emotion ◽  
Clinical Information ◽  
Clinical Decision ◽  
Machine Learning Algorithms ◽  
Schizophrenia Spectrum Disorders ◽  
Real Patient ◽  
Schizophrenia Spectrum ◽  
Spectrum Disorders

AbstractPrior research has identified associations between social media activity and psychiatric diagnoses; however, diagnoses are rarely clinically confirmed. Toward the goal of applying novel approaches to improve outcomes, research using real patient data is necessary. We collected 3,404,959 Facebook messages and 142,390 images across 223 participants (mean age = 23.7; 41.7% male) with schizophrenia spectrum disorders (SSD), mood disorders (MD), and healthy volunteers (HV). We analyzed features uploaded up to 18 months before the first hospitalization using machine learning and built classifiers that distinguished SSD and MD from HV, and SSD from MD. Classification achieved AUC of 0.77 (HV vs. MD), 0.76 (HV vs. SSD), and 0.72 (SSD vs. MD). SSD used more (P < 0.01) perception words (hear, see, feel) than MD or HV. SSD and MD used more (P < 0.01) swear words compared to HV. SSD were more likely to express negative emotions compared to HV (P < 0.01). MD used more words related to biological processes (blood/pain) compared to HV (P < 0.01). The height and width of photos posted by SSD and MD were smaller (P < 0.01) than HV. MD photos contained more blues and less yellows (P < 0.01). Closer to hospitalization, use of punctuation increased (SSD vs HV), use of negative emotion words increased (MD vs. HV), and use of swear words increased (P < 0.01) for SSD and MD compared to HV. Machine-learning algorithms are capable of differentiating SSD and MD using Facebook activity alone over a year in advance of hospitalization. Integrating Facebook data with clinical information could one day serve to inform clinical decision-making.

Effectivness of Health Information System Applications: Clinical Information and Diagnosis-Treatment Systems in Turkey

2017 ◽  
Vol 5 (1) ◽  
pp. 122
Author(s):  
Assist. Prof. Dr. Demokaan DEMİREL
Keyword(s):  
Decision Making ◽  
Information System ◽  
Information Systems ◽  
Management Information System ◽  
Health Sector ◽  
Health Information System ◽  
Clinical Information ◽  
Clinical Decision ◽  
Clinical Information Systems ◽  
Management Information

The distinctive quality of the new social structure is that information becomes the only factor of production. In today's organizations, public administrators are directly responsible for applying information to administrative processes. In addition to his managerial responsibilities, a knowledge based organization requires every employee to take responsibility for achieving efficiency. This has increased the importance of information systems in the decision-making process. Information systems consist of computer and communication technology, data base management and model management and include activity processing system, management information system, decision support systems, senior management information system, expert systems and office automation systems. Information systems in the health sector aim at the management and provision of preventive and curative health services. The use of information systems in healthcare has the benefits of increasing service quality, shortening treatment processes, maximizing efficiency of the time, labour and medical devices. The use of information systems for clinical decision making and reducing medical errors in the healthcare industry dates back to the 1960s. Clinical information systems involve processing, storing and re-accessing information that supports patient care in a hospital. Clinical information systems are systems that are directly or indirectly related to patient care. These systems include electronic health/patient records, clinical decision support systems, nurse information systems, patient tracking systems, tele-medicine, case mix and smart card applications. Diagnosis-treatment systems are information-based systems used in the diagnosis and treatment of diseases. It consists of laboratory information systems, picture archiving and communication system, pharmacy information system, radiology information system, nuclear medicine information system. This study aims to evaluate the effectiveness of health information system applications in Turkey. The first part of the study focuses on the concept of information systems and the types of information systems in organization structures. In the second part, clinical information systems and applications for diagnosis-treatment systems in Turkey are examined. Finally, the study evaluates applications in the health sector qualitatively from the new organizational structure, which is formed by information systems.

A feature-based hybrid recommender system for risk prediction : Machine learning approach (Preprint)

2020 ◽  
Author(s):  
Uzair Bhatti
Keyword(s):  
Machine Learning ◽  
Risk Prediction ◽  
Predictive Accuracy ◽  
Correct Diagnosis ◽  
Recommendation Systems ◽  
Data Integrity ◽  
Machine Learning Algorithms ◽  
Patient Counseling ◽  
Hybrid Filtering ◽  
Novel Algorithm

BACKGROUND In the era of health informatics, exponential growth of information generated by health information systems and healthcare organizations demands expert and intelligent recommendation systems. It has become one of the most valuable tools as it reduces problems such as information overload while selecting and suggesting doctors, hospitals, medicine, diagnosis etc according to patients’ interests. OBJECTIVE Recommendation uses Hybrid Filtering as one of the most popular approaches, but the major limitations of this approach are selectivity and data integrity issues.Mostly existing recommendation systems & risk prediction algorithms focus on a single domain, on the other end cross-domain hybrid filtering is able to alleviate the degree of selectivity and data integrity problems to a better extent. METHODS We propose a novel algorithm for recommendation & predictive model using KNN algorithm with machine learning algorithms and artificial intelligence (AI). We find the factors that directly impact on diseases and propose an approach for predicting the correct diagnosis of different diseases. We have constructed a series of models with good reliability for predicting different surgery complications and identified several novel clinical associations. We proposed a novel algorithm pr-KNN to use KNN for prediction and recommendation of diseases RESULTS Beside that we compared the performance of our algorithm with other machine algorithms and found better performance of our algorithm, with predictive accuracy improving by +3.61%. CONCLUSIONS The potential to directly integrate these predictive tools into EHRs may enable personalized medicine and decision-making at the point of care for patient counseling and as a teaching tool. CLINICALTRIAL dataset for the trials of patient attached

Development of A Drug Early Warning System Model for Cardiac Arrest Using Deep Learning: Retrospective Cohort Study (Preprint)

2020 ◽  
Author(s):  
Hsiao-Ko Chang ◽  
Hui-Chih Wang ◽  
Chih-Fen Huang ◽  
Feipei Lai
Keyword(s):  
Machine Learning ◽  
Time Series ◽  
Cardiac Arrest ◽  
Early Warning ◽  
Time Series Data ◽  
Predictive Accuracy ◽  
Vital Signs ◽  
Warning System ◽  
Series Data ◽  
Dynamic Time

BACKGROUND In most of Taiwan’s medical institutions, congestion is a serious problem for emergency departments. Due to a lack of beds, patients spend more time in emergency retention zones, which make it difficult to detect cardiac arrest (CA). OBJECTIVE We seek to develop a Drug Early Warning System Model (DEWSM), it included drug injections and vital signs as this research important features. We use it to predict cardiac arrest in emergency departments via drug classification and medical expert suggestion. METHODS We propose this new model for detecting cardiac arrest via drug classification and by using a sliding window; we apply learning-based algorithms to time-series data for a DEWSM. By treating drug features as a dynamic time-series factor for cardiopulmonary resuscitation (CPR) patients, we increase sensitivity, reduce false alarm rates and mortality, and increase the model’s accuracy. To evaluate the proposed model, we use the area under the receiver operating characteristic curve (AUROC). RESULTS Four important findings are as follows: (1) We identify the most important drug predictors: bits (intravenous therapy), and replenishers and regulators of water and electrolytes (fluid and electrolyte supplement). The best AUROC of bits is 85%, it means the medical expert suggest the drug features: bits, it will affect the vital signs, and then the evaluate this model correctly classified patients with CPR reach 85%; that of replenishers and regulators of water and electrolytes is 86%. These two features are the most influential of the drug features in the task. (2) We verify feature selection, in which accounting for drugs improve the accuracy: In Task 1, the best AUROC of vital signs is 77%, and that of all features is 86%. In Task 2, the best AUROC of all features is 85%, which demonstrates that thus accounting for the drugs significantly affects prediction. (3) We use a better model: For traditional machine learning, this study adds a new AI technology: the long short-term memory (LSTM) model with the best time-series accuracy, comparable to the traditional random forest (RF) model; the two AUROC measures are 85%. It can be seen that the use of new AI technology will achieve better results, currently comparable to the accuracy of traditional common RF, and the LSTM model can be adjusted in the future to obtain better results. (4) We determine whether the event can be predicted beforehand: The best classifier is still an RF model, in which the observational starting time is 4 hours before the CPR event. Although the accuracy is impaired, the predictive accuracy still reaches 70%. Therefore, we believe that CPR events can be predicted four hours before the event. CONCLUSIONS This paper uses a sliding window to account for dynamic time-series data consisting of the patient’s vital signs and drug injections. The National Early Warning Score (NEWS) only focuses on the score of vital signs, and does not include factors related to drug injections. In this study, the experimental results of adding the drug injections are better than only vital signs. In a comparison with NEWS, we improve predictive accuracy via feature selection, which includes drugs as features. In addition, we use traditional machine learning methods and deep learning (using LSTM method as the main processing time series data) as the basis for comparison of this research. The proposed DEWSM, which offers 4-hour predictions, is better than the NEWS in the literature. This also confirms that the doctor’s heuristic rules are consistent with the results found by machine learning algorithms.

Machine learning in clinical decision making

Med ◽  
2021 ◽  
Author(s):  
Lorenz Adlung ◽  
Yotam Cohen ◽  
Uria Mor ◽  
Eran Elinav
Keyword(s):  
Machine Learning ◽  
Decision Making ◽  
Clinical Decision Making ◽  
Clinical Decision

scholarly journals Machine Learning in Medical Emergencies: a Systematic Review and Analysis

2021 ◽  
Vol 45 (10) ◽  
Author(s):  
Inés Robles Mendo ◽  
Gonçalo Marques ◽  
Isabel de la Torre Díez ◽  
Miguel López-Coronado ◽  
Francisco Martín-Rodríguez
Keyword(s):  
Machine Learning ◽  
Systematic Review ◽  
Language Processing ◽  
Emergency Services ◽  
Clinical Decision ◽  
Quality Of Healthcare ◽  
The Other ◽  
Other Hand ◽  
Increasing Demand ◽  
Commercial Applications

AbstractDespite the increasing demand for artificial intelligence research in medicine, the functionalities of his methods in health emergency remain unclear. Therefore, the authors have conducted this systematic review and a global overview study which aims to identify, analyse, and evaluate the research available on different platforms, and its implementations in healthcare emergencies. The methodology applied for the identification and selection of the scientific studies and the different applications consist of two methods. On the one hand, the PRISMA methodology was carried out in Google Scholar, IEEE Xplore, PubMed ScienceDirect, and Scopus. On the other hand, a review of commercial applications found in the best-known commercial platforms (Android and iOS). A total of 20 studies were included in this review. Most of the included studies were of clinical decisions (n = 4, 20%) or medical services or emergency services (n = 4, 20%). Only 2 were focused on m-health (n = 2, 10%). On the other hand, 12 apps were chosen for full testing on different devices. These apps dealt with pre-hospital medical care (n = 3, 25%) or clinical decision support (n = 3, 25%). In total, half of these apps are based on machine learning based on natural language processing. Machine learning is increasingly applicable to healthcare and offers solutions to improve the efficiency and quality of healthcare. With the emergence of mobile health devices and applications that can use data and assess a patient's real-time health, machine learning is a growing trend in the healthcare industry.

Sign in / Sign up

Export Citation Format

Share Document