Bias in artificial intelligence (AI) is a pervasive challenge, often reinforcing systemic inequities in healthcare systems. This paper proposes an innovative framework to repurpose bias in AI, leveraging it as a tool for addressing structural injustices and improving outcomes for underrepresented and marginalized groups. Traditional healthcare algorithms often exhibit racial biases, such as underestimating risks for black patients or failing to detect dark-skinned individuals in diagnostic or safety-critical applications. This paper redefines AI bias as a tool for equity, proposing a framework to correct systemic healthcare disparities. By introducing purpose-driven bias, AI can enhance fairness in diagnostics, safety, and medical interventions. The approach involves bias analysis, diverse data curation, and AI fine-tuning to align with fairness objectives. This framework highlights the potential of “biased AI” to drive more inclusive and equitable healthcare.
Shah et al. (Thu,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: