SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2606.22692 · 2026

VISTA Architect: A graph database-oriented health AI system demonstrated in multidisciplinary tumor boards

Jason Fries, David Wu, Sylvia Plevritis, Tuomo Kiiskinen, Philip Adamson, Timothy Ellis-Caleo, Aaron Fanous, Balasubramanian Narasimhan, Joel Neal, Manuel Rivas

arXiv · PDF · Open in the Atlas

Code that ran

2 of 2 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 2 failures of our instrument; 0 no recorded run), measured 2026-11-22; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2606.22692.json · badge: /api/ran/2606.22692.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
VISTA-Stanford/meds2text community
Papers with Code archive
2 ran · 0 no recorded run
FunctionStatusWhere it lives
load_metadata Ran
ran; our contract draft was wrong, not the code
VISTA-Stanford/meds2text/src/meds2text/textify.py
code served (permissive licence) · get_code("4dc52f703b8c3a0a")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned dict. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import os
import tempfile
import pandas as pd

# Create temporary directory with test metadata files
temp_dir = tempfile.mkdtemp()

# Create provider.csv
provider_data = {
    'provider_id': ['1', '2', '3'],
    'gender_concept_id': ['8532', '8507', '0'],
    'specialty_source_value': ['Cardiology', 'Neurology', ''],
    'year_of_birth': ['1970', '1975', '1980'],
    'care_site_id': ['10', '20', '30']
}
provider_df = pd.DataFrame(provider_data)
provider_df.to_csv(os.path.join(temp_dir, 'provider.csv'), index=False)

# Create payer_plan_period.csv
payer_plan_data = {
    'person_id': ['100', '101', '102'],
    'payer_plan_period_start_date': ['2020-01-01', '2020-06-15', '2021-01-01'],
    'payer_plan_period_end_date': ['2020-12-31', '2021-06-14', '2021-12-31'],
    'payer_source_value': ['Insurance_A', 'Insurance_B', 'Insurance_A']
}
payer_plan_df = pd.DataFrame(payer_plan_data)
payer_plan_df.to_csv(os.path.join(temp_dir, 'payer_plan_period.csv'), index=False)

# Create care_site.csv
care_site_data = {
    'care_site_id': ['10', '20', '30'],
    'care_site_name': ['Hospital A', 'Clinic B', 'Hospital C']
}
care_site_df = pd.DataFrame(care_site_data)
care_site_df.to_csv(os.path.join(temp_dir, 'care_site.csv'), index=False)

ARGS = {
    'path_to_metadata': temp_dir
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
pandas — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
read_metadata_file Ran
ran; our contract draft was wrong, not the code
VISTA-Stanford/meds2text/src/meds2text/textify.py
code served (permissive licence) · get_code("abda22a3f86268d0")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned DataFrame. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import os
import tempfile
import pandas as pd

# Create a temporary directory and a test CSV file
temp_dir = tempfile.mkdtemp()
test_data = pd.DataFrame({
    'col1': [1, 2, 3],
    'col2': ['a', 'b', 'c']
})
test_csv_path = os.path.join(temp_dir, 'test_metadata.csv')
test_data.to_csv(test_csv_path, index=False)

ARGS = {
    'path_to_metadata': temp_dir,
    'base_name': 'test_metadata'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
pandas — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type DataFrame — a DataFrame output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a DataFrame output the recorded ndim 2 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block

Also present in other papers' code

Of 2 samples lifted from this paper's linked repositories, 2 carry a function identity and 0 do not. Of the 2 distinct functions with an identity, 0 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 0 later) and 2 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

We introduce VISTA Architect, a database-oriented AI architecture for integrating large language models (LLMs) with longitudinal electronic health records (EHRs). At ingestion, it transforms complex clinical documentation into a persistent, provenance-linked knowledge graph, eliminating repeated reprocessing of raw records at query time. The architecture has two layers: a source-faithful MEDS Graph preserving granular EHR structure with full provenance, and a clinically abstracted Timeline Object Architecture (TOA) that uses graph-guided LLM extraction to synthesize a concise timeline of deduplicated, temporally coherent clinical events. This addresses key limitations of direct long-context prompting and retrieval-augmented generation (RAG), which often miss temporal relationships and incur high cost and latency from repeated raw-text processing. By precomputing clinical synthesis once, downstream queries access an organized patient state and traverse to source documentation only when detailed verification is needed. We demonstrate the system in multidisciplinary thoracic oncology tumor boards at Stanford Medicine, where precise reconstruction of patient histories is critical. Across 1,180 patients, VISTA Architect achieved 96.4% accuracy (mean 9.75/10) on 15 tumor board-salient variables (17,700 evaluations; 95% CI 96.1-96.7%), surpassing a matched BM25 RAG baseline and recent benchmarks for LLM-based clinical extraction. An agentic interface reduced preparation for a 30-patient held-out cohort to about 2.2 minutes without sacrificing accuracy. While configured here for thoracic oncology, the modular design adapts to other specialties through customizable event definitions, episode structures, and agentic tools; validation beyond thoracic oncology remains future work.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2606.22692")
get_code_for_paper("2606.22692")
have("2606.22692")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2606.22692.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2606.22692.svg)](https://syntology.ai/paper/2606.22692)

Connect an agent — have() is free.