Patient, Place, Prior (P$^3$): What Counts as Personalization in Medical World Models?

Key points

• Longitudinal models forecast how a patient's imaging state evolves, but accuracy does not show whether the patient's observed trajectory drives the prediction.1
• We also propose Cancer JEPA, a one-step model that forecasts frozen representations of future breast dynamic contrast-enhanced MRI examinations during neoadjuvant therapy.1
• In a validation cohort previously used in development, forecast error is lower when the neural correction receives the patient's history rather than another patient's and patient-matched lesion occupancy maps rather than substituted maps.1
• P$^3$ thus separates input use from evidence of patient-specific predictive value beyond a population-level pattern.1

Sources (1)

• [1]Patient, Place, Prior (P$^3$): What Counts as Personalization in Medical World Models?
  arXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 6, 10:45 PM

  We introduce Patient, Place, Prior (P$^3$), an audit asking whether a forecast benefits from the patient's longitudinal imaging history (Patient), benefits from patient-matched externally supplied spatial support (Place), and gains predictive value beyond a population-average prediction under matched support and context (Prior).

  Longitudinal models forecast how a patient's imaging state evolves, but accuracy does not show whether the patient's observed trajectory drives the prediction.

Extractive summary: sentences quoted from the sources.