Evaluate Recall in Disease Screening

Context

MediScan uses a binary classification model to flag patients as high risk for early-stage lung disease based on imaging and clinical features. The clinical team is concerned that the model looks strong on overall accuracy, but too many true cases may be going undetected.

Current Performance

The Problem

The new model improved accuracy and precision, but recall fell from 74% to 61%. In this setting, missing a true positive can delay diagnosis and treatment. The hiring manager wants to assess whether you understand what recall means, how to interpret it alongside other metrics, and when it should be prioritized over accuracy or precision.

Requirements

1. Define recall clearly and compute what 0.61 recall means in this setting.
2. Explain why accuracy alone is misleading given the 8% disease prevalence.
3. Interpret the tradeoff between precision and recall in the current model.
4. Identify when recall should be prioritized for this product.
5. Recommend practical steps to improve recall without causing unacceptable operational burden.

Constraints

• Radiologists can review at most 250 flagged cases per day.
• False negatives are clinically costly because missed disease delays intervention.
• False positives create follow-up imaging costs and patient anxiety.