Predicting Infectious Disease Using Deep Learning and Big Data

Computational Health Engineering Applied to Model Infectious Diseases and Antimicrobial Resistance Spread

Infectious diseases are the primary cause of mortality worldwide. The dangers of infectious disease are compounded with antimicrobial resistance, which remains the greatest concern for human health. Although novel approaches are under investigation, the World Health Organization predicts that by 2050, septicaemia caused by antimicrobial resistant bacteria could result in 10 million deaths per year. One of the main challenges in medical microbiology is to develop novel experimental approaches, which enable a better understanding of bacterial infections and antimicrobial resistance. After the introduction of whole genome sequencing, there was a great improvement in bacterial detection and identification, which also enabled the characterization of virulence factors and antimicrobial resistance genes. Today, the use of in silico experiments jointly with computational and machine learning offer an in depth understanding of systems biology, allowing us to use this knowledge for the prevention, prediction, and control of infectious disease. Herein, the aim of this review is to discuss the latest advances in human health engineering and their applicability in the control of infectious diseases. An in-depth knowledge of host–pathogen–protein interactions, combined with a better understanding of a host’s immune response and bacterial fitness, are key determinants for halting infectious diseases and antimicrobial resistance dissemination.

Estimating the association between antibiotic exposure and colonization with extended-spectrum β-lactamase-producing Gram-negative bacteria using machine learning methods: a multicentre, prospective cohort study

OBJECTIVES

The aim of the study was to measure the impact of antibiotic exposure on the acquisition of colonization with extended-spectrum β-lactamase-producing Gram-negative bacteria (ESBL-GNB) accounting for individual- and group-level confounding using machine-learning methods.

METHODS

Patients hospitalized between September 2010 and June 2013 at six medical and six surgical wards in Italy, Serbia and Romania were screened for ESBL-GNB at hospital admission, discharge, antibiotic start, and after 3, 7, 15 and 30 days. Primary outcomes were the incidence rate and predictive factors of new ESBL-GNB colonization. Random forest algorithm was used to rank antibiotics according to the risk of selection of ESBL-GNB colonization in patients not colonized before starting antibiotics.

RESULTS

We screened 10 034 patients collecting 28 322 rectal swab samples. New ESBL-GNB colonization incidence with and without antibiotic treatment was 22/1000 and 9/1000 exposure-days, respectively. In the adjusted regression analyses, antibiotic exposure (hazard ratio (HR) 2.38; 95% CI 1.29-4.40), age 60-69 years (HR 1.19; 95% CI 1.05-1.34), and spring season (HR 1.25; 95% CI 1.14-1.38) were independently associated with new colonization. Monotherapy ranked higher als combination therapy in promoting ESBL-GNB colonization. Among monotherapy, cephalosporins ranked first followed by tetracycline (second), macrolide (fourth) and cotrimoxazole (seventh). Overall the ranking of cephalosporins was lower when used in combination. Among combinations not including cephalosporins, quinolones plus carbapenems ranked highest (eighth). Among sequential therapies, quinolones ranked highest (tenth) when prescribed within 30 days of therapy with cephalosporins.

CONCLUSIONS

Impact of antibiotics on selecting ESBL-GNB at intestinal level varies if used in monotherapy or combination and according to previous antibiotic exposure. These finding should be explored in future clinical trials on antibiotic stewardship interventions.

CLINICAL TRIAL REGISTRATION

NCT01208519.

A Deep Neural Network-Based Method for Early Detection of Osteoarthritis Using Statistical Data

A large number of people suffer from certain types of osteoarthritis, such as knee, hip, and spine osteoarthritis. A correct prediction of osteoarthritis is an essential step to effectively diagnose and prevent severe osteoarthritis. Osteoarthritis is commonly diagnosed by experts through manual inspection of patients’ medical images, which are usually collected in hospitals. Checking the occurrence of osteoarthritis is somewhat time-consuming for patients. In addition, the current studies are focused on automatically detecting osteoarthritis through image-based deep learning algorithms. This needs patients’ medical images, which requires patients to visit the hospital. However, medical utilization and health behavior information as statistical data are easier to collect and access than medical images. Using indirect statistical data without any medical images to predict the occurrence of diverse forms of OA can have significant impacts on pro-active and preventive medical care. In this study, we used a deep neural network for detecting the occurrence of osteoarthritis using patient’s statistical data of medical utilization and health behavior information. The study was based on 5749 subjects. Principal component analysis with quantile transformer scaling was employed to generate features from the patients’ simple background medical records and identify the occurrence of osteoarthritis. Our experiments showed that the proposed method using deep neural network with scaled PCA resulted in 76.8% of area under the curve (AUC) and minimized the effort to generate features. Hence, this methos can be a promising tool for patients and doctors to prescreen for possible osteoarthritis to reduce health costs and patients’ time in hospitals.

Identification of Diabetic Patients through Clinical and Para-Clinical Features in Mexico: An Approach Using Deep Neural Networks

Diabetes is a chronic and noncommunicable but preventable disease that is affecting the Mexican population at worrying levels, being the first place in prevalence worldwide. Early diabetes detection has become important to prevent other health conditions that involve low organ yield until the patient death. Based on this problem, this work proposes the architecture of an Artificial Neural Network (ANN) for the automated classification of healthy patients from diabetics patients. The analysis was performed used a set of 19 para-clinical features to determine the health status of the patients. The developed model was evaluated through a statistical analysis based on the calculation of the loss function, accuracy, area under the curve (AUC) and receiving operating characteristics (ROC) curve. The results obtained present statistically significant values, with accuracy of 0.94 and AUC values of 0.98. Based on these results, it is possible to conclude that the ANN implemented in this work can classify patients with presence of diabetes from controls with significant accuracy, presenting preliminary results for the development of a diagnostic tool that can be supportive for health specialists.