Predicting Pulmonary Function From the Analysis of Voice: A Machine Learning Approach.

INTRODUCTION: To self-monitor asthma symptoms, existing methods (e.g. peak flow metre, smart spirometer) require special equipment and are not always used by the patients. Voice recording has the potential to generate surrogate measures of lung function and this study aims to apply machine learning approaches to predict lung function and severity of abnormal lung function from recorded voice for asthma patients. METHODS: A threshold-based mechanism was designed to separate speech and breathing from 323 recordings. Features extracted from these were combined with biological factors to predict lung function. Three predictive models were developed using Random Forest (RF), Support Vector Machine (SVM), and linear regression algorithms: (a) regression models to predict lung function, (b) multi-class classification models to predict severity of lung function abnormality, and (c) binary classification models to predict lung function abnormality. Training and test samples were separated (70%:30%, using balanced portioning), features were normalised, 10-fold cross-validation was used and model performances were evaluated on the test samples. RESULTS: The RF-based regression model performed better with the lowest root mean square error of 10·86. To predict severity of lung function impairment, the SVM-based model performed best in multi-class classification (accuracy = 73.20%), whereas the RF-based model performed best in binary classification models for predicting abnormal lung function (accuracy = 85%). CONCLUSION: Our machine learning approaches can predict lung function, from recorded voice files, better than published approaches. This technique could be used to develop future telehealth solutions including smartphone-based applications which have potential to aid decision making and self-monitoring in asthma.

'Hybrid Doctors' Can Fast Track the Evolution of a Sustainable e-Health Ecosystem in Low Resource Contexts: The Sri Lankan Experience.

Although e-health is an area recognized as essential in the rapid development of healthcare systems in low resource contexts, many challenges prevent the emergence of an effective e-health ecosystem. Lack in capacity around health informatics is one of the main challenges. Based on a longitudinal case study gathering data pertaining to a master's program in biomedical informatics in Sri Lanka designed for doctors, in this paper we demonstrate that creating 'hybrid doctors' may be the way forward. We illustrate how hybrid doctors conversant in healthcare and information and communication technology (ICT) are able to facilitate the creation of an e-health ecosystem in a way that it would contribute significantly to the ICT driven healthcare reforms. Through this case study we highlight the importance of multidisciplinarity, participatory design, strategic investments, learning that aligns with developmental needs, networking, gaining legitimacy and re-packaging perspectives on 'health informatics capacity development'.

Strengthening health systems through informatics capacity development among doctors in low-resource contexts: the Sri Lankan experience.

In the process of strengthening health systems, a lack of health-informatics capacity within low- and middle-income country settings is a considerable challenge. Many capacity-development initiatives on health informatics exist, most of which focus on the adoption of eHealth tools by front-line health-care workers. By contrast, there are only a few programmes that focus on empowering medical doctors in low- and middle-income countries to become champions of digital health innovation and adoption. Sri Lanka has a dynamic eHealth ecosystem, resulting largely from the country's community of medical doctors who are also health informaticians. They are the result of a decade-long programme centred on a Master of Science degree course in biomedical informatics, which has trained over 150 medical doctors to date, and has now been extended to a specialist training programme. This paper evaluates this unique capacity-development effort from the perspective of strengthening health systems and how those in other low- and middle-income country contexts may learn from the Sri Lankan experience when implementing capacity-development programmes in health informatics.