Evolution of Machine Learning in Tuberculosis Diagnosis: A Review of Deep Learning-Based Medical Applications

Singh, Manisha; Pujar, Gurubasavaraj V.; Kumar, Sethu Arun; Bhagyalalitha, Meduri; Akshatha, H.S.; Abuhaija, Belal; Alsoud, Anas Ratib; Abualigah, Laith; Beeraka, Narasimha M; Gandomi, Amir H.

doi:10.3390/electronics11172634

Cited by 49 publications

(25 citation statements)

References 118 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Using samples collected between 2005 and 2011, we were able to train classification models to predict which patients go on to form transmission clusters in 2012 and 2013 and those that would likely not lead to further transmission events. Recently, machine learning has been used in TB research to improve disease diagnosis, for example the automated detection of lesions on lung x-rays 10 . Our results suggest that we can use classification models, including machine learning approaches, to classify individuals with TB that represent a higher or lower risk of further transmission through shared demographic and disease features.…”

Section: Discussionmentioning

confidence: 99%

Early prediction ofMycobacterium tuberculosistransmission clusters using supervised learning models

Gharamaleki,

Colijn,

Sekirov

et al. 2024

Preprint

View full text Add to dashboard Cite

Identifying individuals with tuberculosis with a high risk of onward transmission can guide disease prevention and public health strategies. Here, we train classification models to predict the first sampled isolates in Mycobacterium tuberculosis transmission clusters from demographic and disease data. We find that supervised learning models, in particular balanced random forests, can be used to develop predictive models that discriminate between individuals with TB that are more likely to form transmission clusters and individuals that are likely not to transmit further, with good model performance and AUCs of more than 0.75. We also identified the most important patient and disease characteristics in the best performing classification model, including patient demographics, site of infection, TB lineage, and age at diagnosis. This framework can be used to develop predictive tools for the early assessment of a patient's transmission risk to prioritise individuals for enhanced follow-up with the aim of reducing further transmission.

show abstract

Section: Discussionmentioning

confidence: 99%

Early prediction ofMycobacterium tuberculosistransmission clusters using supervised learning models

Gharamaleki,

Colijn,

Sekirov

et al. 2024

Preprint

View full text Add to dashboard Cite

show abstract

“…With the development of artificial intelligence technology, machine learning can help us to develop new and easy diagnostic or predictive tools in clinical work, based on population-based big data. It has been applied in many common diseases, such as tuberculosis [ 20 ] and Alzheimer's disease [ 21 ]. Furthermore, some physician scientists use machine deep learning to pioneer the next generation of medical robotics [ 22 ].…”

Section: Discussionmentioning

confidence: 99%

Risk Evaluation of Bone Metastases and a Simple Tool for Detecting Bone Metastases in Prostate Cancer: A Population‐Based Study

Chen²,

Zhu

et al. 2023

Computational and Mathematical Methods in Medicine

View full text Add to dashboard Cite

Introduction. Population-based estimates of the incidence and prognosis of bone metastases in prostate cancer (PC) are lacking. We aimed to characterize the incidence and risk of bone metastases and develop a simple tool for the prediction of bone metastases among patients with PC. Methods. Data were obtained from the Surveillance, Epidemiology, and End Results (SEER) database. A total of 75698 patients with PC with confirmed presence or absence of bone metastases at diagnosis between 1975 and 2019 in the United States were used for analysis. Data were stratified by age, race, residence, median income, prostate-specific antigen (PSA) values, tumor size, distant metastatic history, and positive lymph node scores. Multivariable logistic and Cox regressions were performed to identify predictors of bone metastases and factors correlated with all-cause mortality. Classification tree analysis was performed to establish a model. Results. After patients with PC with missing data were excluded, 75698 cases remained. Among these, 3835 patients had bone metastases. Incidence proportions were highest in patients with a high prostate-specific antigen (PSA) value (odds ratio (OR), 2.49; 95% confidence interval (CI), 1.35-4.35; p < 0.002 ). Multivariable Cox regression and risk analyses indicated that high PSA values (hazards ratio (HR), 19.8; 95% CI, 18.5-21.2; p < 0.001 ) and high positive lymph node scores (vs. score 0; HR, 8.65; 95% CI, 7.89-9.49; p < 0.001 ) were significant risk factors for mortality. Meanwhile, in the predication tree analysis, PSA values and lymph node scores were the most significant determining factors in two models. Median survival among the patients with PC was 78 months, but only 31 months among those with bone metastases. Conclusion. Patients with PC with high PSA values or high positive lymph node scores were at a significantly higher risk of bone metastases. Our study may provide a simple and accurate tool to identify patients with PC at high risk of bone metastases based on population-based estimates.

show abstract

“…The AUC of train model and valid model are both greater than 0.9. Compared with several literatures ( Manisha Singh et al, 2022 ; Yu Dong Zhang, 2020 ) which utilize the chest radiography images to detect Pulmonary Tuberculosis with the help of machine learning tools (CAD, DL, ICNN), our work is analysing biomarkers from peripheral blood biomarkers and constructing diagnostic model for PTB with the combination of RF and ANN. Although, RF, ANN, or other machine learning had been utilized in diagnosing TB ( Dande and Samant, 2018 ; Orjuela-Canon et al, 2022 ), combining RF and ANN to diagnose PTB had never been reported.…”

Section: Discussionmentioning

confidence: 99%

A united model for diagnosing pulmonary tuberculosis with random forest and artificial neural network

Zhu

Liu

2023

Front. Genet.

View full text Add to dashboard Cite

Background: Pulmonary tuberculosis (PTB) is a chronic infectious disease and is the most common type of TB. Although the sputum smear test is a gold standard for diagnosing PTB, the method has numerous limitations, including low sensitivity, low specificity, and insufficient samples.Methods: The present study aimed to identify specific biomarkers of PTB and construct a model for diagnosing PTB by combining random forest (RF) and artificial neural network (ANN) algorithms. Two publicly available cohorts of TB, namely, the GSE83456 (training) and GSE42834 (validation) cohorts, were retrieved from the Gene Expression Omnibus (GEO) database. A total of 45 and 61 differentially expressed genes (DEGs) were identified between the PTB and control samples, respectively, by screening the GSE83456 cohort. An RF classifier was used for identifying specific biomarkers, following which an ANN-based classification model was constructed for identifying PTB samples. The accuracy of the ANN model was validated using the receiver operating characteristic (ROC) curve. The proportion of 22 types of immunocytes in the PTB samples was measured using the CIBERSORT algorithm, and the correlations between the immunocytes were determined.Results: Differential analysis revealed that 11 and 22 DEGs were upregulated and downregulated, respectively, and 11 biomarkers specific to PTB were identified by the RF classifier. The weights of these biomarkers were determined and an ANN-based classification model was subsequently constructed. The model exhibited outstanding performance, as revealed by the area under the curve (AUC), which was 1.000 for the training cohort. The AUC of the validation cohort was 0.946, which further confirmed the accuracy of the model.Conclusion: Altogether, the present study successfully identified specific genetic biomarkers of PTB and constructed a highly accurate model for the diagnosis of PTB based on blood samples. The model developed herein can serve as a reliable reference for the early detection of PTB and provide novel perspectives into the pathogenesis of PTB.

show abstract

Evolution of Machine Learning in Tuberculosis Diagnosis: A Review of Deep Learning-Based Medical Applications

Cited by 49 publications

References 118 publications

Early prediction ofMycobacterium tuberculosistransmission clusters using supervised learning models

Early prediction ofMycobacterium tuberculosistransmission clusters using supervised learning models

Risk Evaluation of Bone Metastases and a Simple Tool for Detecting Bone Metastases in Prostate Cancer: A Population‐Based Study

A united model for diagnosing pulmonary tuberculosis with random forest and artificial neural network

Contact Info

Product

Resources

About