Preserve the protocol identity. The original Pasquier result is thirteen component scores summed to 0-39; a single global 0-3 impression is a labeled later adaptation.
{
"categories": [
{
"outcome_code": "0",
"sulcal_morphology": "normal_gyri_and_no_sulcal_opening",
"ventricular_morphology": "no_enlargement"
},
{
"outcome_code": "1",
"sulcal_morphology": "opening_of_sulci_without_clear_gyral_volume_loss",
"ventricular_morphology": "mild_enlargement"
},
{
"outcome_code": "2",
"sulcal_morphology": "gyral_volume_loss_with_wider_sulci",
"ventricular_morphology": "moderate_enlargement"
},
{
"outcome_code": "3",
"sulcal_morphology": "severe_knife_blade_gyral_loss",
"ventricular_morphology": "severe_enlargement"
}
],
"identity_and_applicability": {
"original_name": "Pasquier_cerebral_atrophy_assessment",
"original_population": "Fifty patients aged 19-81 years with acute hemispheric ischemic stroke, rated by four observers twice 24 hours apart.",
"original_modality_and_plane": "Axial T2-weighted MRI.",
"use_for": "Structured visual description of cerebral sulcal and ventricular atrophy burden with regional and side-specific preservation.",
"do_not_use_for": [
"standalone_dementia_diagnosis",
"Alzheimer_biomarker_status",
"hydrocephalus_diagnosis",
"individual_conversion_probability",
"treatment_eligibility"
]
},
"original_13_region_architecture": {
"total_region_count": 13,
"warning": "Thirteen regions total, not thirteen per hemisphere.",
"regional_scores": [
{
"id": "right_frontal_sulci",
"side": "right",
"structure": "frontal_sulci",
"allowed_codes": [
"0",
"1",
"2",
"3"
]
},
{
"id": "left_frontal_sulci",
"side": "left",
"structure": "frontal_sulci",
"allowed_codes": [
"0",
"1",
"2",
"3"
]
},
{
"id": "right_parieto_occipital_sulci",
"side": "right",
"structure": "parieto_occipital_sulci",
"allowed_codes": [
"0",
"1",
"2",
"3"
]
},
{
"id": "left_parieto_occipital_sulci",
"side": "left",
"structure": "parieto_occipital_sulci",
"allowed_codes": [
"0",
"1",
"2",
"3"
]
},
{
"id": "right_temporal_sulci",
"side": "right",
"structure": "temporal_sulci",
"allowed_codes": [
"0",
"1",
"2",
"3"
]
},
{
"id": "left_temporal_sulci",
"side": "left",
"structure": "temporal_sulci",
"allowed_codes": [
"0",
"1",
"2",
"3"
]
},
{
"id": "right_frontal_horn",
"side": "right",
"structure": "lateral_ventricle_frontal_horn",
"allowed_codes": [
"0",
"1",
"2",
"3"
]
},
{
"id": "left_frontal_horn",
"side": "left",
"structure": "lateral_ventricle_frontal_horn",
"allowed_codes": [
"0",
"1",
"2",
"3"
]
},
{
"id": "right_occipital_horn",
"side": "right",
"structure": "lateral_ventricle_occipital_horn",
"allowed_codes": [
"0",
"1",
"2",
"3"
]
},
{
"id": "left_occipital_horn",
"side": "left",
"structure": "lateral_ventricle_occipital_horn",
"allowed_codes": [
"0",
"1",
"2",
"3"
]
},
{
"id": "right_temporal_horn",
"side": "right",
"structure": "lateral_ventricle_temporal_horn",
"allowed_codes": [
"0",
"1",
"2",
"3"
]
},
{
"id": "left_temporal_horn",
"side": "left",
"structure": "lateral_ventricle_temporal_horn",
"allowed_codes": [
"0",
"1",
"2",
"3"
]
},
{
"id": "third_ventricle",
"side": "midline",
"structure": "third_ventricle",
"allowed_codes": [
"0",
"1",
"2",
"3"
]
}
],
"calculation": "Sum all thirteen regional scores. Minimum 0 and maximum 39; retain every component so the total is reproducible.",
"output": "original_regional_total_0_to_39_plus_13_component_scores"
},
"protocol_modes_and_noninterchangeability": {
"original_regional_mode": "Thirteen separate 0-3 ratings summed to 0-39, as published by Pasquier et al. in 1996.",
"simplified_global_mode": "A single overall 0-3 visual impression used in later dementia-imaging practice. It may use the same morphology words but is not the original 0-39 sum.",
"guard": "Always return protocol_mode. Never label one overall 0-3 impression as the original Pasquier total, never multiply it by 13, and never reconstruct regional values from it."
},
"imaging_protocol_and_quality": {
"original_sequence": "Axial T2 MRI.",
"current_practice": "Use a consistent MRI sequence and plane, commonly FLAIR or T1 for visual atrophy assessment, or explicitly label a CT adaptation. Compare only like protocols when judging longitudinal change.",
"confounders": [
"head_rotation_or_obliquity",
"motion",
"slice_thickness",
"different_modality_or_sequence",
"large_infarct_or_encephalomalacia",
"mass_effect",
"postoperative_change",
"hydrocephalus_pattern"
],
"focal_lesion_guard": "Do not convert focal tissue loss from infarction, trauma, surgery or another destructive lesion into generalized neurodegenerative atrophy without describing the cause and distribution."
},
"pattern_and_differential": {
"normal_aging_pattern": "Age-associated volume loss is generally more symmetric and generalized; lobar, focal or asymmetric atrophy requires separate localization and etiologic context.",
"complementary_scales": "Report medial temporal atrophy with Scheltens MTA, posterior atrophy with Koedam and WMH burden with the two-axis Fazekas scale when relevant; these scores are complementary, not interchangeable.",
"hydrocephalus_boundary": "Ventricular enlargement may be ex vacuo or reflect another CSF-dynamics pattern. Assess sulci, callosal angle, temporal horns, flow-related findings and clinical context rather than diagnosing hydrocephalus from a GCA grade."
},
"reliability_and_longitudinal_limits": {
"original_results": "In the 1996 study, complete interobserver agreement was 41.7% in the first session and 44.1% in the second; mean overall interobserver kappa was 0.48 then 0.67, and mean intraobserver kappa was 0.65.",
"author_boundary": "The original authors considered the method reproducible when performed by one observer; retain rater and protocol for serial comparison.",
"CT_MRI_boundary": "Modern CT-versus-MRI reliability work supports structured use but does not make scores across modalities automatically interchangeable for subtle or longitudinal change.",
"sensitivity_guard": "GCA is a coarse visual scale and is not sufficiently sensitive to quantify small short-interval volume change."
},
"current_interpretation_and_care_context": {
"age_context": "Later dementia-imaging teaching treats simplified grade 3 as abnormal at any age and grade 2 as concerning below about 75 years, but these are rule-of-thumb contextual thresholds, not universal diagnostic cutoffs.",
"symptom_context": "When moderate or severe, focal or asymmetric atrophy is unexpected for age or concordant with cognitive or neurologic symptoms, integrate formal clinical assessment, history, examination, laboratory testing and disease-specific imaging or biomarkers as appropriate.",
"Alzheimer_boundary": "GCA cannot establish amyloid or tau biology, diagnose Alzheimer disease on its own, stage a patient biologically or determine eligibility for disease-modifying treatment.",
"grade_only_guard": "No 0-3 regional grade, 0-39 original total or simplified global score automatically orders medication, biomarker testing, follow-up interval or specialist referral."
},
"risk_and_interpretation": {
"no_calibrated_probability": "The system provides no validated per-grade probability of dementia, Alzheimer pathology, conversion, functional decline or survival for an individual.",
"necessary_covariates": [
"age",
"education_and_baseline_function",
"symptoms_and_cognitive_profile",
"regional_pattern_and_asymmetry",
"longitudinal_change",
"vascular_and_other_imaging_findings",
"relevant_biomarkers"
],
"grade_0_guard": "A normal or minimal GCA appearance does not exclude early, focal or biomarker-defined neurodegenerative disease."
},
"agent_output_contract": [
"protocol_mode_original_13_region_or_simplified_global",
"modality_sequence_plane_and_quality",
"thirteen_regional_scores_and_total_0_to_39_when_original",
"global_0_to_3_with_adaptation_label_when_simplified",
"right_left_asymmetry_and_lobar_pattern",
"focal_destructive_lesions_and_hydrocephalus_differential",
"age_symptom_and_comparison_context",
"complementary_MTA_Koedam_and_Fazekas_findings",
"no_standalone_diagnosis_risk_or_treatment_warning",
"uncertainty_and_missing_inputs"
],
"missing_input_behavior": [
"If only a single 0-3 value is provided, store it as simplified global mode unless thirteen regional inputs prove the original sum was performed.",
"If fewer than thirteen original regions are assessable, return the available components and an incomplete total rather than imputing missing scores.",
"If modality, sequence or side is unknown, do not claim longitudinal progression or protocol-equivalent scoring.",
"If atrophy is focal, asymmetric or confounded by a destructive lesion, describe that pattern and avoid an unqualified generalized label."
],
"evidence_map": [
{
"role": "original_scale_and_reliability",
"citation": "Pasquier et al., European Neurology 1996",
"doi": "10.1159/000117270"
},
{
"role": "simplified_global_practice_context",
"citation": "Vernooij and van Buchem, Neuroimaging in Dementia, NCBI Bookshelf",
"url": "https://www.ncbi.nlm.nih.gov/books/NBK554327/"
},
{
"role": "CT_MRI_reliability",
"citation": "Hobden et al., Neurological Sciences 2024",
"doi": "10.1007/s10072-023-07113-z"
},
{
"role": "Alzheimer_biologic_boundary",
"citation": "Alzheimer's Association revised criteria 2024",
"doi": "10.1002/alz.13859"
}
],
"source_locator": "Pasquier et al., Eur Neurol 1996;36:268-272, PMID 8864706, abstract and methods/results for the 13-region 0-39 MRI architecture and reproducibility; Vernooij and van Buchem, NCBI Bookshelf NBK554327, Fig 11.3 and Table 11.1 for simplified 0-3 morphology and age-context cautions; Hobden et al. 2024, PMC10942897, Table 2 and reliability results for CT versus MRI; Jack et al. 2024, DOI 10.1002/alz.13859, biomarker-based Alzheimer diagnostic boundary."
}