Decision Support
❯
Diagnose
❯Prediction of Multidrug-Resistant TB from CT Pulmonary Images Based on Deep Learning Techniques
Large Companies
Knowledge representation - Bayesian Network
For:
Process Engineers.
Scope:
Applying Bayesian Networks to root cause analysis of industrial processes.
Goal:
Improve Operation Efficiency
Large Companies
Operations - Intelligent Monitoring of Infrastructures
For:
Plant Operators
Scope:
Optimization of thermal efficiency of a power plant using Machine Learning.
Goal:
Improve Operation Efficiency, Improved Employee Efficiency, Reduce costs
General Public
Healthcare - Improve Investigatory Work
For:
Radiologists
Scope:
Using machine learning to predict lung cancer from CT scans.
Goal:
Anticipate Risks
Prediction of Multidrug-Resistant TB from CT Pulmonary Images Based on Deep Learning Techniques
For:
Clinicians and medical professionals predicting multidrug-resistant (MDR) patients from drug-sensitive (DS) ones based on CT lung images.
Goal:
Improved Employee Efficiency
Prediction of Multidrug-Resistant TB from CT Pulmonary Images Based on Deep Learning Techniques
For:
Clinicians and medical professionals predicting multidrug-resistant (MDR) patients from drug-sensitive (DS) ones based on CT lung images.Goal:
Improved Employee EfficiencyProblem addressed
Some versions, or abnormities, of tuberculosis are resistant to multiple drugs used to treat this disease/infection. Although these abnormities can sometimes be observed and referenced to in CT Pulmonary scans by professional clinicians (doctors), they are sometimes not obvious.
The conventional diagnostic procedure comprises applying microbiological culture that takes several weeks and remains expensive.
Description
To complement the conventional diagnostic procedure, high resolution computer tomography (CT) of pulmonary images has been resorted to not only for aiding clinicians to expedite the process of diagnosis but also for monitoring prognosis when administering antibiotic drugs.
As a result, the proposed architecture of CNN + SVM + patch performs the best with classification accuracy rate at 91.11%, on these image scans.
Image
AI: Perceive
Machine Learning
AI: Understand
Diagnose
AI: Act