Context

MediScan, a telehealth platform, uses a binary classification model to flag chest X-rays for possible pneumonia so radiologists can prioritize urgent cases. The team is debating whether the current model is acceptable because leadership sees high precision, while clinicians are worried about missed positive cases.

Current Performance

Confusion Matrix (Current Model)

The product manager argues the model is strong because 91% of flagged scans are truly positive. The clinical lead argues the model is unsafe because it misses 380 of 1,000 pneumonia cases.

Task

1. Explain the difference between precision and recall using the metrics above.
2. Interpret what the current values mean for this medical triage use case.
3. Compare the current model with the baseline and explain which is preferable under different business or clinical priorities.
4. Describe how changing the decision threshold would likely affect precision and recall.
5. Recommend what metric the team should optimize for and why.

Constraints

• Missing a true pneumonia case delays treatment and has high clinical risk.
• Too many false positives increase radiologist workload and queue time.
• The review team can absorb at most 20% more flagged scans without hiring.