Your team developed a machine learning tool for a hospital. The tool takes X$-$ray images as inputs, and for each image input, it outputs the decision on whether or not lung cancer is detected from the X$-$ray. You learned from the hospital that $95\%$ of people who took the $x-$ray do not have lung cancer. If a patient took multiple $x-$ray exams, only the most recent one would appear in the dataset. Your model achieved $90\%$ accuracy, meaning that $90\%$ of the decisions your model made are correct. Which of the following statements is incorrect?

In addition to accuracy, it might be beneficial to evaluate the precision $($proportion of true cancer in the predicted cancer$)$ and recall $($proportion of true cancer identified$)$

The machine learning model you developed is not very effective in the accuracy aspect, because a naive model that predicts all X$-$ray images to be cancer$-$free will have $95\%$ accuracy, higher than the $90\%$ your model achieved.

The machine learning model you developed is very effective, because it is correct $90\%$ of the time, which is a very high number.