scholarly journals Explainable Machine Learning Prediction for COVID-19 Mortality in the Colombian Population

Author(s):  
Gabriel Ricardo Vásquez Morales ◽  
Sergio Mauricio Martínez Monterrubio ◽  
Juan Antonio Recio García ◽  
Pablo Moreno Ger

Abstract The COVID-19 pandemic, which began in late 2019, has become a global public health problem, resulting in large numbers of infections and deaths. One of the greatest challenges in dealing with the disease is to identify those people who are most at risk of becoming infected, seriously ill and dying from the virus so that they can be isolated in a targeted manner to reduce mortality rates. This article proposes using machine learning, specifically neural networks, and random forests, to build two complementary models that identify the probability that a person has of dying because of COVID-19. The models are trained with the demographic information and medical history of two population groups: 43,000 people who died from COVID-19 in Colombia during 2020, and a random sample of 43,000 people who became ill with COVID-19 during the same period but later recovered. After training the neural network classification model, evaluation metrics were applied that yielded an 88% accuracy value. However, transparency is a major requirement for the explicability of the COVID-19 prognosis. Therefore, a complementary random forest model was trained that identified the most significant predictors of mortality by COVID-19.

2021 ◽  
Author(s):  
Gabriel Ricardo Vásquez Morales ◽  
Sergio Mauricio Martínez Monterrubio ◽  
Juan Antonio Recio García ◽  
Pablo Moreno Ger

Abstract The COVID-19 pandemic, which began in late 2019, has become a global public health problem, resulting in large numbers of people infected and dead. One of the greatest challenges in dealing with the disease is to identify those people who are most at risk of becoming infected, seriously ill and dying from the virus, so that they can be isolated in a targeted manner and thus reduce mortality rates. This article proposes the use of machine learning, and specifically of neural networks and random forest to build two complementary models that identify the probability that a person has of dying because of COVID-19. The models are trained with the demographic information and medical history of two population groups: on the one hand, 43,000 people who died from COVID-19 in Colombia during 2020, and on the other hand, a random sample of 43,000 people who became ill with COVID-19 during the same period of time, but later recovered. After training the neural network classification model, evaluation metrics were applied that yielded an 88% accuracy value. However, transparency is a major requirement for the explicability of COVID-19 prognosis. Therefore, a complementary random forest model is trained that allows the identification of the most significant predictors of mortality by COVID-19.


2021 ◽  
Author(s):  
Ambreen Tharani ◽  
Salima Farooq ◽  
Maryam Ali ◽  
Uroosa Talib ◽  
Murad Moosa Khan

Abstract Background: Self-Harm (SH) is a major global public health problem which is under-researched in Pakistan. A prior act of self-harm is one of the strongest predictors of future suicide.Method: This retrospective descriptive study describes the characteristics of SH cases (n=350) that presented to a tertiary care teaching hospital in Karachi, Pakistan, from January 2013 to December 2017. Details related to demography, history, associated factors, access to methods used, and intent to die were collected on a structured proforma and analysed using STATA version 14. Results: It was found that self-harm acts were twice as more common in females than in males. More than half of the reported cases were in the age group 20-39 years. Drug overdose and use of insecticides were the two most common methods used in both genders. Depression was identified in nearly half of the reported SH cases. Intention to die was found to be 3 times greater among patients with psychiatric illness as compared to those with no history of psychiatric illness. Conclusion: This study suggests that limiting access to lethal means, regulating over-the-counter sale of medications, and safe storage of pesticides can possibly serve as effective measures to minimize self-harm incidences. Moreover, integration of suicide assessment and prevention programmes for the general population is also suggested.


2021 ◽  
Vol 34 (1) ◽  
pp. e100347
Author(s):  
Tosin Philip Oyetunji ◽  
S M Yasir Arafat ◽  
Stephen Oluwaseyi Famori ◽  
Timilehin Blessing Akinboyewa ◽  
Michael Afolami ◽  
...  

BackgroundSuicide is a global public health problem and Nigeria is one of the epicentres of suicide in the world. However, there is a dearth of research exploring the epidemiological aspects of suicide in Nigeria.AimTo examine the demographic information and precipitating events for suicides in Nigeria by analysing the contents of newspaper reports of suicide.MethodsWe searched, collected, and analysed published news reports about suicide from 10 English newspapers in Nigeria. A total of 350 suicide reports were assessed between January 2010 and December 2019 after screening and sorting.ResultsThe mean (SD) age of the reported cases was 36.33 (15.48) years. Majority of the reported cases were male (80.6%), married (51.8%), students (33.6%), living in a semi-urban area (40.3%) and among the age group of 25–34 (25.3%). Hanging (48.6%) and poisoning (32.2%) were the most commonly reported methods of suicide. Financial constraints and marital conflicts were most commonly assumed precipitating factors.ConclusionThis study suggests that being male, married, or living in semi-urban areas are associated with suicide in Nigeria. Further community-based studies are warranted to generalise the findings and adopt appropriate preventive strategies.


2019 ◽  
Vol 19 (8) ◽  
pp. 567-578 ◽  
Author(s):  
Marcus Vinicius Nora de Souza ◽  
Thais Cristina Mendonça Nogueira

Nowadays, tuberculosis (TB) is an important global public health problem, being responsible for millions of TB-related deaths worldwide. Due to the increased number of cases and resistance of Mycobacterium tuberculosis to all drugs used for the treatment of this disease, we desperately need new drugs and strategies that could reduce treatment time with fewer side effects, reduced cost and highly active drugs against resistant strains and latent disease. Considering that, 4H-1,3-benzothiazin-4-one is a promising class of antimycobacterial agents in special against TB-resistant strains being the aim of this review the discussion of different aspects of this chemical class such as synthesis, mechanism of action, medicinal chemistry and combination with other drugs.


Dermatology ◽  
2021 ◽  
pp. 1-7
Author(s):  
Irene Torrecilla-Martínez ◽  
Esperanza Manrique-Silva ◽  
Victor Traves ◽  
Celia Requena ◽  
Eduardo Nagore

<b><i>Background:</i></b> The incidence of cutaneous melanoma, an important global public health problem, has been increasing over the last several decades. <b><i>Objectives:</i></b> In order to decrease melanoma-related mortality, ways to communicate and implement the correct methods for conducting primary and secondary prevention measures (such as early detection via self-examination) should be investigated. <b><i>Materials and Methods:</i></b> An observational, cross-sectional, retrospective study consisting of 409 patients diagnosed with cutaneous melanoma was conducted. An online questionnaire was created to evaluate knowledge levels, attitudes, and adherence to primary preventive measures and to skin self-examination practices. <b><i>Results:</i></b> The results revealed that even when 43% of the patients perform cutaneous self-examinations, only half of them fully followed the recommendations. Patients aged &#x3c;45 years, female, with a I–II phototype, with an intermediate/high level of education, and with a history of NMSC were more likely to have an adequate degree of knowledge. Moreover, patients aged &#x3c;45 years and with an adequate degree of knowledge more frequently showed an adequate adherence to the primary prevention measures. Finally, patients aged 45–60 years and with an adequate degree of knowledge presented a good adherence to the self-skin examination measures. <b><i>Limitations:</i></b> Possible limitations of this study were memory bias through the influence of age within the study population, and bias due to a greater proportion of subjects with a high education level. <b><i>Conclusion:</i></b> Within the population of patients with melanoma, a high percentage of patients do not rigorously follow the recommended prevention measures. Our study highlights the need to implement awareness in this population to improve the prevention of cutaneous cancer.


Author(s):  
VEERENDRA UPPARA ◽  
SAISEKHAR KODIVANDLA ◽  
ASHIK ALI SHAIK

Heart failure (HF) is a major global public health problem irrespective of its causes. It generates an enormous clinical, societal, and economic, health loss burden with an increase in its prevalence reaching an epidemic proportion. The morbidity and mortality associated with heart failure are increasing the health-related burdens worldwide, especially in low- and middle-income countries. This review highlights the trends in HF burden, the clinical spectrum of HF, and the importance of neurohormonal pathways and the evolution of angiotensin receptor neprilysin inhibition in HF with updated clinical practice guidelines.


2016 ◽  
Vol 27 (1) ◽  
pp. 31-35
Author(s):  
Montosh Kumar Mondal ◽  
Beauty Rani Roy ◽  
Shibani Banik ◽  
Debabrata Banik

Medication error is a major cause of morbidity and mortality in medical profession . There is an increasing recognition that medication errors are causing a substantial global public health problem, as many result in harm to patients and increased costs to health providers.Anaesthesia is now safe and routine, yet anaesthetists are not immune from making medication errors and the consequences of their mistakes may be more serious than those of doctors in other specialties. Steps are being taken to determine the extent of the problem of medication error in anaesthesia. In this review, incidence, types, risk factors and preventive measures of the medication errors are discussed in detail.Journal of Bangladesh Society of Anaesthesiologists 2014; 27(1): 31-35


Sign in / Sign up

Export Citation Format

Share Document