New Delhi: Scientists have developed a new artificial intelligence (AI) model to predict treatment outcomes for tuberculosis (TB) patients, a breakthrough that may lead to personalized treatments for bacterial diseases. The study, published in the journal iScience, analyzed multimodal data including various biomedical data from clinical trials, genomics, medical imaging and drug prescriptions from TB patients. By analyzing data from patients with varying levels of drug resistance, researchers discovered biomedical characteristics that predict treatment failure.
They also discovered effective drug regimens against specific groups of patients with drug-resistant tuberculosis. “Our multimodal AI model accurately predicted treatment prognosis and outperformed existing models that focus on a limited set of clinical data,” said Sriram Chandrasekaran, corresponding author and associate professor at the University of Michigan, US. “We identified drug regimens that were effective against certain types of drug-resistant tuberculosis in all countries, which is very important due to the spread of drug-resistant tuberculosis,” added the study’s first author, Awanti Sambarey, a fellow postdoctoral fellow at the University of Michigan.
Using AI, the team examined more than 5,000 patients. “We’re talking about real-world data, so patients from different countries have different admission protocols. We work with more than 200 biomedical characteristics in our analysis; we examine demographic information such as age and gender, as well as treatment history “said Sambarey. “We also looked at whether patients had other comorbidities, such as HIV, and then worked with various imaging features such as X-rays, CT scans, pathogen data, drug resistance data, as well as genomic features and what mutations they had. the pathogen,” he said.
The researchers noted that clinically it is very difficult to analyze the data in its entirety. That is where the role of AI comes in handy. The team also studied the impact of the type of drug resistance present. “You can look at a specific snapshot of the data, such as genomic characteristics, and find what mutations the infecting pathogen had, and ask what some of the long-term treatment implications are,” Sambarey added.
The researchers found that certain drug combinations worked better in patients with some types of resistance but not others, leading to treatment failure. They also found that medications with antagonistic pharmaceutical interactions could lead to worse outcomes. “Using AI to eliminate antagonistic drugs early in the drug discovery process can prevent treatment failure in the future,” Chandrasekaran said. “Rather than a one-size-fits-all treatment approach, we hope that studying multimodal data will help clinicians treat patients with more personalized treatments to provide the best outcomes,” Sambarey added.