Nishani Gunawardana
Research Subjects
Nishani Gunawardana has a total of 21 co-authors. Here are the top co-authors by number of shared publications:
- Ravindra Prasan Rannan-Eliya
Institute for Health Policy, Colombo, Sri Lanka
3 shared publications - Anuji Gamage
General Sir John Kotelawala Defence University
2 shared publications - Chathurani Sigera
Faculty of Medicine, University of Colombo, 25, Kynsey Road, Colombo, 00800, Sri Lanka
2 shared publications - Herath Mudiyanselage Meththananada Herath
University of Ruhuna
2 shared publications - Ishwari Sivagnanam
Institute for Health Policy, Colombo, Sri Lanka
2 shared publications - Nethmi Weerawardena
Institute for Health Policy, Colombo, Sri Lanka
2 shared publications
Latest Publications
- Open Access
- 1 July 2024
- Published by SAGE Publications in Health Informatics Journal
- Vol. 30 (3)
Machine learning approaches for asthma disease prediction among adults in Sri Lanka
Objectives: Addressing the challenge of cost-effective asthma diagnosis amidst diverse symptom patterns among patients, this study aims to develop a machine learning-based asthma prediction tool for self-detection of asthma. Methods: Data from 6,665 participants in the Sri Lanka Health and Ageing Study (2018-2019) are used for this research. Thirteen machine learning algorithms, including Logistic Regression, Support Vector Machine, Decision Tree, Random Forest, Naïve Bayes, K-Nearest Neighbors, Gradient Boost, XGBoost, AdaBoost, CatBoost, LightGBM, Multi-Layer Perceptron, and Probabilistic Neural Network, are employed. Results: A hybrid version of Logistic Regression and LightGBM outperformed other models, achieving an AUC of 0.9062 and 79.85% sensitivity. Key predictive features for asthma include wheezing, breathlessness with wheezing, shortness of breath attacks, coughing attacks, chest tightness, nasal allergies, physical activity, passive smoking, ethnicity, and residential sector. Conclusion: Combining Logistic Regression and LightGBM models can effectively predict adult asthma based on self-reported symptoms and demographic and behavioural characteristics. The proposed expert system assists clinicians and patients in diagnosing potential asthma cases.
- 1 July 2024
- Published by SAGE Publications in Health Informatics Journal
- Vol. 30 (3)
Open AccessMachine learning approaches for asthma disease prediction among adults in Sri Lanka
Objectives: Addressing the challenge of cost-effective asthma diagnosis amidst diverse symptom patterns among patients, this study aims to develop a machine learning-based asthma prediction tool for self-detection of asthma. Methods: Data from 6,665 participants in the Sri Lanka Health and Ageing Study (2018-2019) are used for this research. Thirteen machine learning algorithms, including Logistic Regression, Support Vector Machine, Decision Tree, Random Forest, Naïve Bayes, K-Nearest Neighbors, Gradient Boost, XGBoost, AdaBoost, CatBoost, LightGBM, Multi-Layer Perceptron, and Probabilistic Neural Network, are employed. Results: A hybrid version of Logistic Regression and LightGBM outperformed other models, achieving an AUC of 0.9062 and 79.85% sensitivity. Key predictive features for asthma include wheezing, breathlessness with wheezing, shortness of breath attacks, coughing attacks, chest tightness, nasal allergies, physical activity, passive smoking, ethnicity, and residential sector. Conclusion: Combining Logistic Regression and LightGBM models can effectively predict adult asthma based on self-reported symptoms and demographic and behavioural characteristics. The proposed expert system assists clinicians and patients in diagnosing potential asthma cases.
- Open Access
- 15 February 2023
- Published by BMJ in BMJ Open Diabetes Research & Care
- Vol. 11 (1) , e003160
Prevalence of diabetes and pre-diabetes in Sri Lanka: a new global hotspot–estimates from the Sri Lanka Health and Ageing Survey 2018/2019
Introduction This study’s objective was to produce robust, comparable estimates of the prevalence of diabetes and pre-diabetes in the Sri Lankan adult population, where previous studies suggest the highest prevalence in South Asia. Research design and methods We used data on 6661 adults from the nationally representative 2018/2019 first wave of the Sri Lanka Health and Ageing Study (SLHAS). We classified glycemic status based on previous diabetes diagnosis, and either fasting plasma glucose (FPG), or FPG and 2-hour plasma glucose (2-h PG). We estimated crude and age-standardized prevalence of pre-diabetes and diabetes and by major individual characteristics weighting the data to account for study design and subject participation. Results Crude prevalence of diabetes in adults was 23.0% (95% CI 21.2% to 24.7%) using both 2-h PG and FPG, and age-standardized prevalence was 21.8% (95% CI 20.1% to 23.5%). Using only FPG, prevalence was 18.5% (95% CI 7.1% to 19.8%). Previously diagnosed prevalence was 14.3% (95% CI 13.1% to 15.5%) of all adults. The prevalence of pre-diabetes was 30.5% (95% CI 28.2% to 32.7%). Diabetes prevalence increased with age until ages ≥70 years and was more prevalent in female, urban, more affluent, and Muslim adults. Diabetes and pre-diabetes prevalence increased with body mass index (BMI) but was as high as 21% and 29%, respectively, in those of normal weight. Conclusions Study limitations included using only a single visit to assess diabetes, relying on self-reported fasting times, and unavailability of glycated hemoglobin for most participants. Our results indicate that Sri Lanka has a very high diabetes prevalence, significantly higher than previous estimates of 8%–15% and higher than current global estimates for any other Asian country. Our results have implications for other populations of South Asian origin, and the high prevalence of diabetes and dysglycemia at normal body weight indicates the need for further research to understand the underlying drivers.
- 15 February 2023
- Published by BMJ in BMJ Open Diabetes Research & Care
- Vol. 11 (1) , e003160
Open AccessPrevalence of diabetes and pre-diabetes in Sri Lanka: a new global hotspot–estimates from the Sri Lanka Health and Ageing Survey 2018/2019
Introduction This study’s objective was to produce robust, comparable estimates of the prevalence of diabetes and pre-diabetes in the Sri Lankan adult population, where previous studies suggest the highest prevalence in South Asia. Research design and methods We used data on 6661 adults from the nationally representative 2018/2019 first wave of the Sri Lanka Health and Ageing Study (SLHAS). We classified glycemic status based on previous diabetes diagnosis, and either fasting plasma glucose (FPG), or FPG and 2-hour plasma glucose (2-h PG). We estimated crude and age-standardized prevalence of pre-diabetes and diabetes and by major individual characteristics weighting the data to account for study design and subject participation. Results Crude prevalence of diabetes in adults was 23.0% (95% CI 21.2% to 24.7%) using both 2-h PG and FPG, and age-standardized prevalence was 21.8% (95% CI 20.1% to 23.5%). Using only FPG, prevalence was 18.5% (95% CI 7.1% to 19.8%). Previously diagnosed prevalence was 14.3% (95% CI 13.1% to 15.5%) of all adults. The prevalence of pre-diabetes was 30.5% (95% CI 28.2% to 32.7%). Diabetes prevalence increased with age until ages ≥70 years and was more prevalent in female, urban, more affluent, and Muslim adults. Diabetes and pre-diabetes prevalence increased with body mass index (BMI) but was as high as 21% and 29%, respectively, in those of normal weight. Conclusions Study limitations included using only a single visit to assess diabetes, relying on self-reported fasting times, and unavailability of glycated hemoglobin for most participants. Our results indicate that Sri Lanka has a very high diabetes prevalence, significantly higher than previous estimates of 8%–15% and higher than current global estimates for any other Asian country. Our results have implications for other populations of South Asian origin, and the high prevalence of diabetes and dysglycemia at normal body weight indicates the need for further research to understand the underlying drivers.
- Open Access
- 11 January 2022
- Published by Ubiquity Press, Ltd. in Global Heart
- Vol. 17 (1) , 50
Prevalence and Associations of Hypertension in Sri Lankan Adults: Estimates from the SLHAS 2018–19 Survey Using JNC7 and ACC/AHA 2017 Guidelines
Sri Lanka lacks robust estimates of hypertension (HTN) prevalence owing to few national studies, hindering optimization of control strategies. Evidence on how the revised 2017 American College of Cardiology/American Heart Association (ACC/AHA) HTN definition affects prevalence in low- and middle-income countries (LMICs) is also limited. To make robust estimates of HTN prevalence in the Sri Lankan adult population, and to assess impact of the ACC/AHA 2017 definitions. Data were sourced from the 2018–2019 first wave of the Sri Lanka Health and Ageing Study (SLHAS), a nationally representative longitudinal study of the noninstitutionalized adult population. After excluding those with missing data and aged <18 years, 6,342 participants (95.1%) were included in the analysis. HTN was defined using either the traditional threshold of systolic BP (SBP) ≥140 mmHg or a diastolic BP (DBP) ≥90 mmHg, or the ACC/AHA 2017 threshold of SBP ≥130 mmHg or DBP ≥80 mmHg, or if taking antihypertensive medication. Estimated prevalence of HTN in all Sri Lankan adults was 28.2% using the traditional definition, and it doubled to 51.3% when applying the ACC/AHA 2017 definition. Of those classified as hypertensive according to the older and ACC/AHA 2017 definitions, 53.4% and 31.2%, respectively, were previously diagnosed. Of the 23.2% of adults reclassified as hypertensive by the ACC/AHA 2017 definition, 16.6% had a history of CVD or diabetes. Increased prevalence was associated with urban residence, socioeconomic status, obesity, and Muslim ethnicity. Prevalence increased with age, but the increase was steeper in women from their 30s. Nearly one in three adult Sri Lankans are hypertensive, requiring antihypertensive treatment. Applying the ACC/AHA 2017 definitions almost doubles numbers, but many of those reclassified would require treatment under recent WHO guidelines. Study findings also suggest that design effects in HTN surveys may be higher than usually assumed.
- 11 January 2022
- Published by Ubiquity Press, Ltd. in Global Heart
- Vol. 17 (1) , 50
Open AccessPrevalence and Associations of Hypertension in Sri Lankan Adults: Estimates from the SLHAS 2018–19 Survey Using JNC7 and ACC/AHA 2017 Guidelines
Sri Lanka lacks robust estimates of hypertension (HTN) prevalence owing to few national studies, hindering optimization of control strategies. Evidence on how the revised 2017 American College of Cardiology/American Heart Association (ACC/AHA) HTN definition affects prevalence in low- and middle-income countries (LMICs) is also limited. To make robust estimates of HTN prevalence in the Sri Lankan adult population, and to assess impact of the ACC/AHA 2017 definitions. Data were sourced from the 2018–2019 first wave of the Sri Lanka Health and Ageing Study (SLHAS), a nationally representative longitudinal study of the noninstitutionalized adult population. After excluding those with missing data and aged <18 years, 6,342 participants (95.1%) were included in the analysis. HTN was defined using either the traditional threshold of systolic BP (SBP) ≥140 mmHg or a diastolic BP (DBP) ≥90 mmHg, or the ACC/AHA 2017 threshold of SBP ≥130 mmHg or DBP ≥80 mmHg, or if taking antihypertensive medication. Estimated prevalence of HTN in all Sri Lankan adults was 28.2% using the traditional definition, and it doubled to 51.3% when applying the ACC/AHA 2017 definition. Of those classified as hypertensive according to the older and ACC/AHA 2017 definitions, 53.4% and 31.2%, respectively, were previously diagnosed. Of the 23.2% of adults reclassified as hypertensive by the ACC/AHA 2017 definition, 16.6% had a history of CVD or diabetes. Increased prevalence was associated with urban residence, socioeconomic status, obesity, and Muslim ethnicity. Prevalence increased with age, but the increase was steeper in women from their 30s. Nearly one in three adult Sri Lankans are hypertensive, requiring antihypertensive treatment. Applying the ACC/AHA 2017 definitions almost doubles numbers, but many of those reclassified would require treatment under recent WHO guidelines. Study findings also suggest that design effects in HTN surveys may be higher than usually assumed.
- Open Access
- 21 November 2019
- Published by Sri Lanka Journals Online in Journal of Science of the University of Kelaniya
- Vol. 12, 45-65
Factors associated with induction of labour and pregnancy outcomes in 14 healthcare facilities in Sri Lanka
Induction of Labour (IOL) is an important practice that is carried out commonly in modern day obstetrics. In medium to large healthcare facilities in Sri Lanka, it is estimated that approximately 35.5% of all deliveries involve IOL. This research attempts to identify the factors that affect IOL and to assess the association between IOL and the pregnancy outcome. In this study, we considered 18309 women who were admitted to 14 healthcare facilities for delivery in 3 randomly selected provinces in Sri Lanka (Western, Southern and Eastern provinces), during July to October 2011. Multinomial Logistic Regression model (MLR) and Fuzzy Expert System (FES) were used to identify the factors that lead to IOL. The MLR model predicts IOL with a classification rate of 65.5% and the FES predicts IOL with an accuracy of 55.10%. Maternal age, number of previous caesarian sections, number of previous births, estimated gestational age, Pre-Eclampsia, number of previous pregnancies, Placenta Preavia, Abruption Placenta, total number of neonates delivered, birth weight and Maternal Severity Index (MSI) were identified as factors associated with IOL. Neonatal status after seven days of life can also be predicted using the developed FES. FES is predictive of IOL and birth outcome, where if the FES score is between 0.8570 and 0.8854, the patient will belong to the induced group and the baby would be alive after seven days of birth. This study concludes that, MLR and FES models can be used to predict IOL outcomes. These findings can be informative to healthcare providers when counselling women for labour induction and develop evidence-based protocols on IOL.
- 21 November 2019
- Published by Sri Lanka Journals Online in Journal of Science of the University of Kelaniya
- Vol. 12, 45-65
Open AccessFactors associated with induction of labour and pregnancy outcomes in 14 healthcare facilities in Sri Lanka
Induction of Labour (IOL) is an important practice that is carried out commonly in modern day obstetrics. In medium to large healthcare facilities in Sri Lanka, it is estimated that approximately 35.5% of all deliveries involve IOL. This research attempts to identify the factors that affect IOL and to assess the association between IOL and the pregnancy outcome. In this study, we considered 18309 women who were admitted to 14 healthcare facilities for delivery in 3 randomly selected provinces in Sri Lanka (Western, Southern and Eastern provinces), during July to October 2011. Multinomial Logistic Regression model (MLR) and Fuzzy Expert System (FES) were used to identify the factors that lead to IOL. The MLR model predicts IOL with a classification rate of 65.5% and the FES predicts IOL with an accuracy of 55.10%. Maternal age, number of previous caesarian sections, number of previous births, estimated gestational age, Pre-Eclampsia, number of previous pregnancies, Placenta Preavia, Abruption Placenta, total number of neonates delivered, birth weight and Maternal Severity Index (MSI) were identified as factors associated with IOL. Neonatal status after seven days of life can also be predicted using the developed FES. FES is predictive of IOL and birth outcome, where if the FES score is between 0.8570 and 0.8854, the patient will belong to the induced group and the baby would be alive after seven days of birth. This study concludes that, MLR and FES models can be used to predict IOL outcomes. These findings can be informative to healthcare providers when counselling women for labour induction and develop evidence-based protocols on IOL.