Tuberculosis is a serious health threat, especially for people living with HIV. People living with HIV are more likely than others to become sick with TB. Worldwide, TB is one of the leading causes of death among people living with HIV.Without treatment, as with other opportunistic infections, HIV and TB can work together to shorten lifespan. Someone with untreated latent TB infection and HIV infection is much more likely to develop TB disease during his or her lifetime than someone without HIV infection.
Among people with latent TB infection, HIV infection is the strongest known risk factor for progressing to TB disease. A person who has both HIV infection and TB disease has an AIDS-defining condition. People infected with HIV who also have either latent TB infection or TB disease can be effectively treated. The first step is to ensure that people living with HIV are tested for TB infection. If found to have TB infection, further tests are needed to rule out TB disease.
Tuberculosis remains a global health problem with an enormous burden of disease, estimated at 10.4 million new cases in 2015. To stop the tuberculosis epidemic, it is critical that we interrupt tuberculosis transmission. Further, the interventions required to interrupt tuberculosis transmission must be targeted to high-risk groups and settings.
The global rate of decline in tuberculosis incidence is currently 1.5% and will need to increase to 4%–5% by 2020 and then to 10% per year by 2025 to meet the World Health Organization End TB Strategy targets. Interrupting tuberculosis transmission is central to achieving the reductions in tuberculosis incidence required to meet the End TB targets. A rate of decline of 10% per year is thought to be achievable, as this was observed during the 1950s and 1960s in Western Europe, where comprehensive tuberculosis control efforts, which included infection control and treatment of M. tuberculosis infection and all forms of tuberculosis, were intensified and universal health coverage and socioeconomic development were expanded.
The infectiousness and duration thereof for a person with tuberculosis depend on host and bacterial factors. Persons with smear-positive pulmonary tuberculosis are highly infectious, and the degree of infectiousness is thought to increase with the degree of smear positivity. In a large study of household contacts in Peru, smear-positive index cases were associated with a higher risk of infection among household contacts, compared with smear-negative index cases, regardless of the age of the household contacts. Persons with smear-negative tuberculosis cases may, however, also transmit tuberculosis. Nevertheless, scale-up of sputum smear microscopy has not succeeded in achieving dramatic declines in tuberculosis incidence. Possible reasons for the lack of impact include the poor sensitivity of smear microscopy, particularly among HIV-infected persons and children, and the occurrence of many cases of transmission before people receive a tuberculosis diagnosis and treatment.
Persons with active pulmonary or laryngeal tuberculosis generate droplet nuclei that contain M. tuberculosis through coughing, singing, shouting, sneezing, or any other forceful expiratory maneuver that shears respiratory secretions from the airways, with coughing being the most efficient at generating infectious aerosols. Appropriate treatment of individuals with infectious tuberculosis results in a rapid reduction in infectiousness. Individuals with index tuberculosis cases who are HIV infected, particularly those with advanced immunosuppression, were hypothesized to be less likely than HIV-uninfected individuals with tuberculosis to transmit to household contacts, possibly because of a greater likelihood of having smear-negative tuberculosis and a shorter duration of infectiousness due to more rapid progression to death.
XGBoost is an open-source software library which provides a regularizing gradient boosting framework for C++, Java, Python, R, Julia, Perl, and Scala. It works on Linux, Windows, and macOS. From the project description, it aims to provide a "Scalable, Portable and Distributed Gradient Boosting Library". The basic idea of the XGBoost algorithm is to take the decision tree as the primary classifier, use the change of loss function to control the complexity of the tree, and combine multiple tree models through continuous iteration to obtain a model with higher accuracy finally. The eXtreme Gradient Boosting (XGBoost) model is used to classify tuberculosis and normal cases.