Tuberculose E Covid-19: Aspectos Clínicos, Prevenção E Controle No Ambiente Prisional

Deon, Reges Antonio; Santos, Paula Cristina dos; Feitosa, Samuel da Silva; Espinoza, Jean Marcel de Almeida; Korb, Arnildo

doi:10.22533/at.ed.6322108128

Cited by 1 publication

(1 citation statement)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, it is observed that not every patient exhibiting symptoms indicative of TB receives appropriate treatment or even a definitive diagnosis of the condition. This issue is particularly pronounced in prison complexes 5 , where TB screening remains suboptimal.…”

Section: Introductionmentioning

confidence: 99%

Artificial intelligence model in Body Scan images for monitoring tuberculosis in a prison complex

Deon,

Korb,

Santos

2024

JHS

View full text Add to dashboard Cite

Tuberculosis (TB) remains a significant public health challenge worldwide, with its transmission exacerbated by various risk factors including co-existing health conditions and socio-economic determinants such as high population density, poverty, and alcoholism. This study emphasizes the crucial role of efficient TB screening and monitoring in not only providing prompt treatment to patients but also in reducing the disease's lethality. Responding to the Ministry of Health and the World Health Organization's demand for advanced diagnostic methods, we introduce a novel approach using the Marie. AI model for TB screening in penitentiary complexes. This proof of value (POV) repurposes Body Scan images, typically used for object detection on inmates, as a new tool for health screening. The Marie.AI model, a multimodal artificial intelligence platform developed in 2020, has previously proven effective in Brazil for aiding healthcare teams in diagnosing COVID-19 and TB. For this study, the model was trained on an extensive dataset of 1.5 million images, including X-rays and CT scans of patients with TB, COVID-19, and other pulmonary diseases, along with patient symptom data. A significant achievement of this study was the model's ability to distinguish between TB-infected and non-infected individuals using Body Scan images, leveraging its previous training with X-ray and CT data. The model demonstrated exceptional diagnostic accuracy, achieving a specificity of 87.23% and a sensitivity of 100% in identifying suspected TB cases. These findings not only highlight the versatility of the Marie.AI model in non-traditional settings but also mark a breakthrough in early TB diagnosis, particularly in high-risk environments like penitentiary complexes. This innovative approach promises to enhance public health responses to TB, leading to more effective disease management and control.

show abstract

Section: Introductionmentioning

confidence: 99%