1
|
Nijiati M, Guo L, Tuersun A, Damola M, Abulizi A, Dong J, Xia L, Hong K, Zou X. Deep learning on longitudinal CT scans: automated prediction of treatment outcomes in hospitalized tuberculosis patients. iScience 2023; 26:108326. [PMID: 37965132 PMCID: PMC10641748 DOI: 10.1016/j.isci.2023.108326] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/14/2023] [Revised: 08/17/2023] [Accepted: 10/20/2023] [Indexed: 11/16/2023] Open
Abstract
Three deep learning (DL)-based prediction models (PMs) using longitudinal CT images were developed to predict tuberculosis (TB) treatment outcomes. The internal dataset consists of 493 bacteriologically confirmed TB patients who completed the anti-tuberculosis treatment with three-time CT scans, including a pretreatment CT scan and two follow-up CT scans. PM1 was trained using only pretreatment CT scans, and PM2 and PM3 were developed by adding follow-up scans. An independent testing was performed on external dataset comprising 86 TB patients. The area under the curve for classifying success and drug-resistant (DR)-TB was improved on both internal (0.609 vs. 0.625 vs. 0.815) and external (0.627 vs. 0.705 vs. 0.735) dataset by adding follow-up scans. The accuracy and F1-score also showed an increasing tendency in the external test. Regular follow-up CT scans can aid in the treatment prediction, and special attention should be given to early intensive phase of treatment to identify high-risk DR-TB patients.
Collapse
Affiliation(s)
- Mayidili Nijiati
- Department of Radiology, The First People’s Hospital of Kashi Prefecture, Kashi, China
| | - Lin Guo
- Shenzhen Zhiying Medical Imaging, Shenzhen, China
| | - Abudouresuli Tuersun
- Department of Radiology, The First People’s Hospital of Kashi Prefecture, Kashi, China
| | - Maihemitijiang Damola
- Department of Radiology, The First People’s Hospital of Kashi Prefecture, Kashi, China
| | | | - Jiake Dong
- Department of Radiology, The First People’s Hospital of Kashi Prefecture, Kashi, China
| | - Li Xia
- Shenzhen Zhiying Medical Imaging, Shenzhen, China
| | - Kunlei Hong
- Shenzhen Zhiying Medical Imaging, Shenzhen, China
| | - Xiaoguang Zou
- Clinical Medical Research Center, The First People’s Hospital of Kashi Prefecture, Kashi, China
| |
Collapse
|
2
|
Laamarti M, El Fathi Lalaoui Y, Elfermi R, Daoud R, El Allali A. Afro-TB dataset as a large scale genomic data of Mycobacterium tuberuclosis in Africa. Sci Data 2023; 10:212. [PMID: 37059737 PMCID: PMC10102689 DOI: 10.1038/s41597-023-02112-3] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/03/2023] [Accepted: 03/28/2023] [Indexed: 04/16/2023] Open
Abstract
Mycobacterium tuberculosis (MTB) is a pathogenic bacterium accountable for 10.6 million new infections with tuberculosis (TB) in 2021. The fact that the genetic sequences of M. tuberculosis vary widely provides a basis for understanding how this bacterium causes disease, how the immune system responds to it, how it has evolved over time, and how it is distributed geographically. However, despite extensive research efforts, the evolution and transmission of MTB in Africa remain poorly understood. In this study, we used 17,641 strains from 26 countries to create the first curated African Mycobacterium tuberculosis (MTB) classification and resistance dataset, containing 13,753 strains. We identified 157 mutations in 12 genes associated with resistance and additional new mutations potentially associated with resistance. The resistance profile was used to classify strains. We also performed a phylogenetic classification of each isolate and prepared the data in a format that can be used for phylogenetic and comparative analysis of tuberculosis worldwide. These genomic data will extend current information for comparative genomic studies to understand the mechanisms and evolution of MTB drug resistance.
Collapse
Affiliation(s)
- Meriem Laamarti
- African Genome Center, Mohammed VI Polytechnic University, Ben Guerir, 43150, Morocco.
| | | | - Rachid Elfermi
- African Genome Center, Mohammed VI Polytechnic University, Ben Guerir, 43150, Morocco
| | - Rachid Daoud
- African Genome Center, Mohammed VI Polytechnic University, Ben Guerir, 43150, Morocco.
| | - Achraf El Allali
- African Genome Center, Mohammed VI Polytechnic University, Ben Guerir, 43150, Morocco.
| |
Collapse
|
3
|
Saavedra Cervera B, López MG, Chiner-Oms Á, García AM, Cancino-Muñoz I, Torres-Puente M, Villamayor L, Madrazo-Moya C, Mambuque E, Sequera GV, Respeito D, Blanco S, Augusto O, López-Varela E, García-Basteiro AL, Comas I. Fine-grain population structure and transmission patterns of Mycobacterium tuberculosis in southern Mozambique, a high TB/HIV burden area. Microb Genom 2022; 8. [PMID: 35787782 PMCID: PMC9455694 DOI: 10.1099/mgen.0.000844] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.5] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 12/30/2022] Open
Abstract
Genomic studies of the Mycobacterium tuberculosis complex (MTBC) might shed light on the dynamics of its transmission, especially in high-burden settings, where recent outbreaks are embedded in the complex natural history of the disease. To this end, we conducted a 1 year prospective surveillance-based study in Mozambique. We applied whole-genome sequencing (WGS) to 295 positive cultures. We fully characterized MTBC isolates by phylogenetics and dating evaluation, and carried out a molecular epidemiology analysis to investigate further associations with pre-defined transmission risk factors. The majority of strains (49.5%, 136/275) belonged to lineage (L) 4; 57.8 % of them (159/275) were in genomic transmission clusters (cut-off 5 SNPs), and a strikingly high proportion (45.5%) shared an identical genotype (0 SNP pairwise distance). We found two ‘likely endemic’ clades, comprising 67 strains, belonging to L1.2, which dated back to the late 19th century and were associated with recent spread among people living with human immunodeficiency virus (PLHIV). We describe for the first time the population structure of MTBC in our region, a high tuberculosis (TB)/HIV burden area. Clustering analysis revealed an unforeseen pattern of spread and high rates of progression to active TB, suggesting weaknesses in TB control activities. The long-term presence of local strains in Mozambique, which were responsible for large transmission among HIV/TB-coinfected patients, calls into question the role of HIV in TB transmission.
Collapse
Affiliation(s)
- Belén Saavedra Cervera
- PhD Programin Medicine and Translational Research, Universitat de Barcelona, Barcelona, Spain.,Centro de Investigação em Saúde de Manhiça (CISM), Maputo, Mozambique.,ISGlobal, Hospital Clínic - Universitat de Barcelona, Barcelona, Spain
| | - Mariana G López
- Instituto de Biomedicina de Valencia (IBV), CSIC, Valencia, Spain
| | | | - Ana María García
- Instituto de Biomedicina de Valencia (IBV), CSIC, Valencia, Spain.,Universidad de Valencia, Valencia, Spain
| | | | | | | | | | - Edson Mambuque
- Centro de Investigação em Saúde de Manhiça (CISM), Maputo, Mozambique
| | | | - Durval Respeito
- Centro de Investigação em Saúde de Manhiça (CISM), Maputo, Mozambique
| | - Silvia Blanco
- Centro de Investigação em Saúde de Manhiça (CISM), Maputo, Mozambique
| | - Orvalho Augusto
- Centro de Investigação em Saúde de Manhiça (CISM), Maputo, Mozambique
| | - Elisa López-Varela
- Centro de Investigação em Saúde de Manhiça (CISM), Maputo, Mozambique.,ISGlobal, Hospital Clínic - Universitat de Barcelona, Barcelona, Spain
| | - Alberto L García-Basteiro
- Centro de Investigação em Saúde de Manhiça (CISM), Maputo, Mozambique.,ISGlobal, Hospital Clínic - Universitat de Barcelona, Barcelona, Spain.,Centro de Investigación Biomédica en Red de Enfermedades Infecciosas (CIBERINFEC), Barcelona, Spain
| | - Iñaki Comas
- Instituto de Biomedicina de Valencia (IBV), CSIC, Valencia, Spain.,CIBER in Epidemiology and Public Health, Madrid, Spain
| |
Collapse
|