Skip to main content
Back to challenges
Challenge

Alzheimer's Blood Test Risk Scoring Engine with LlamaIndex

NIH news details breakthroughs in novel Alzheimer's blood tests (e.g., p-tau217) that predict when cognitive symptoms are on the horizon. In this challenge, developers will build an intelligent clinical decision support engine using LlamaIndex. The system will index patient biomarker assay histories, clinical risk factor literature, and EHR notes to output early Alzheimer's symptom onset timeline estimates.

Data ScienceHosted by Vera
Sign in free to start
Challenge brief

What you are building

The core problem, expected build, and operating context for this challenge.

Build an Alzheimer's blood test predictive risk engine using LlamaIndex structured query index tools.

Delivery guide

How work is evaluated

Evaluation

Evaluates risk scoring accuracy and timeline estimation based on plasma biomarker ratios.

Datasets

Challenge datasets

Public datasets, plus the private datasets you have access to.

Evaluation rubric

How submissions are scored

These dimensions define what the evaluator checks and which criteria separate a passable run from a strong one.

Dimensions
2 scoring checks
Binary
2 pass or fail dimensions
Ordinal
0 scaled dimensions
Dimension 1

valid_risk_tier

Risk tier must be LOW, MED, or HIGH

Binary check

This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.

Dimension 2

risk_score_mae

Mean Absolute Error of calculated risk score against validated longitudinal clinical cohort outcome • target: 0.05 • range: 0-1

Binary check

This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.

Learning goals

What you should walk away with

  • Construct LlamaIndex structured indices over patient blood biomarker records

  • Integrate plasma p-tau217, Aβ42/40 ratio, and Neurofilament Light (NfL) metrics

  • Query prospective cognitive decline timing using clinical literature RAG nodes

  • Generate clinician-facing decision support summaries with citation traceability

Resources and assets

Reference links and supporting material

Dataset notes

Synthetic plasma biomarker profiles (p-tau217, Aβ42/Aβ40, NfL) paired with longitudinal cognitive diagnostic evaluations.

Challenge setup

Evaluates risk scoring accuracy and timeline estimation based on plasma biomarker ratios.

Challenge input

JSON containing p_tau217_pg_ml, abeta_42_40_ratio, age

LlamaIndex

Mandatory LlamaIndex framework

Evaluated output

JSON containing risk_score (float 0-1), projected_onset_years (float), risk_tier (LOW/MED/HIGH)

Evaluation checks
  • Risk tier must be LOW, MED, or HIGH
  • Mean Absolute Error of calculated risk score against validated longitudinal clinical cohort outcome • targe...
Configured evaluation evidence
  • Benchmark: AD-Biomarker-Bench
  • Risk Score Mae target: 0.05
  • 1 public reference case
Configured runtime
  • Python execution harness
  • Python sandbox (unavailable on Versalist)

Summary shows the first task, up to three tools, and up to three evidence items. Open the technical recipe to view all configured items.

View technical recipe

Configured tools

Action Space
  • Llama Index · Required
  • HiddenLayer · Optional
Policy Serving
  • Mistral Saba · Optional

Evaluation contract

  • valid_risk_tier · Weight 1
  • risk_score_mae · Weight 1

Metrics

  • risk_score_mae · Target 0.05

Benchmarks

  • AD-Biomarker-Bench
  • biotech

Reference cases

  • 1 public reference case

Workflow tasks

  1. Task 1

    Input: JSON containing p_tau217_pg_ml, abeta_42_40_ratio, age

    Output: JSON containing risk_score (float 0-1), projected_onset_years (float), risk_tier (LOW/MED/HIGH)

    Execution harness: Python

Evaluation tools

  • LlamaIndex · Agent Framework

Recipe freeze marker

No recipe freeze time is recorded.

Frequently Asked Questions about Alzheimer's Blood Test Risk Scoring Engine with LlamaIndex