Even the NYTimes writes that the evidence for "RF Brain" scans are questionable

>The scientific basis for this technology, which reads brain waves, is scant. Results from BEOS tests are inadmissible as evidence in Indian courts, but the technology has been applied in hundreds of high-stakes cases, including those of murder, rape and terrorism.

People who worked with EEGs (what is probably meant here with RF brain scans) know that getting any good data from EEG measurements is cumbersome. There are currently no known neural correlates suitable for lie detection that could be measured with EEG, and even if they were, the robustness and reliability of them would be highly questionable.

The trouble is that people have multiple contradictory thoughts, and they even think of things which they themselves disagree with. There is a vast difference between what people think and what they do.