Database Credentialed Access

MIMIC-IV-Ext MIMIC Note Extract: A Large-Scale Structured Dataset Extracted from MIMIC-IV-Note Discharge Summaries

Jonas Kuntzer ,  Kristian Schwethelm ,  Conrad Ketzer ,  Eva Wende ,  Moritz Schlager ,  Daniel Rueckert ,  Georgios Kaissis ,  Friederike Jungmann ,  Paul Hager

Published Oct. 21, 2026 · Version 1.0
When using this resource, please cite:

Kuntzer, J., Schwethelm, K., Ketzer, C., Wende, E., Schlager, M., Rueckert, D., Kaissis, G., Jungmann, F., & Hager, P. (2026). MIMIC-IV-Ext MIMIC Note Extract: A Large-Scale Structured Dataset Extracted from MIMIC-IV-Note Discharge Summaries (version 1.0). PhysioNet. RRID:SCR_007345. https://doi.org/10.13026/cq50-rd11

Please include the standard citation for PhysioNet:

Pollard, T., Moody, B. E., Lehman, L., Gow, B., Fernandes, C., Xie, C., Johnson, A., Mark, R. G., & Heldt, T. (2026). PhysioNet as a global platform for biomedical research. Nature Health. https://doi.org/10.1038/s44360-026-00096-z. Available from: https://rdcu.be/faatM

Abstract

MIMIC is one of the most widely used public hospital datasets for machine learning on electronic health records [3], but its structured tables answer many clinical questions only through coded surrogates, or not at all. For example, a patient's past medical history, documented allergies, and pertinent diagnostic findings are only available in MIMIC's discharge summaries. Studies that need this information must therefore extract it from unstructured text, a labor-intensive process that is repeated independently across projects.

We release MIMIC-Note-Extract (MNE), a large-scale dataset containing data from all 16 sections of 300,720 MIMIC-IV-Note v2.2 discharge summaries. We use regex rules and large language models (LLMs) to transform these sections into structured tables while preserving their clinical content. Validation by three medical doctors confirms extraction accuracy above 92% across sections.

Background

The structured tables of MIMIC-IV [1] are missing much of the narrative context of patient care, which appear only in unstructured MIMIC-IV-Note [2] discharge summaries. This leads to a fragmented ecosystem of individual preprocessing where, in the absence of a standardized pipeline, each project must build its own scripts to extract the specific variables needed for its use case. These isolated preprocessing pipelines are rarely shared or unified, resulting in duplicated engineering effort and inconsistent extraction quality across the community. But even if such an effort were made, the resulting datasets are typically tailored so closely to individual projects that they lack the generality needed to support broader research questions. At the core of this lies a methodological limitation: most existing projects focus on information extraction. Instead, we posit that the community requires information conversion; preserving the underlying information of the raw data while transforming it into an easily queryable, structured format.

We introduce MIMIC-Note-Extract (MNE), a structured dataset derived from 300,720 MIMIC-IV-Note discharge summaries. We employ a hybrid processing framework that pairs deterministic parsing with schema-guided LLM generation to translate 16 major clinical sections of the discharge summary into structured tables. By providing structured representations for 13 of these sections, spanning pre-admission history to detailed physical examinations, MNE makes the information directly accessible for downstream use.

Methods

Cohort construction

We derive the cohort from the 331,793 raw discharge summaries in MIMIC-IV-Note v2.2 [2]. Quality filtering removes non-discharge note types, formatting outliers such as specific external dictation system templates, and reports with disordered or severely incomplete clinical sections. The final dataset comprises 300,720 hospital admissions from 137,402 patients.

Splitting into Sections

We identified 24 distinct clinical sections that appear in consistent relative order across MIMIC-IV-Note discharge summaries. Eight of these are entirely removed by upstream de-identification, for example, Name, Admission Date, and Date of Birth are dropped, leaving 16 sections.

Splitting is complicated by partially redacted headers. For example, the header "Discharge ___" may indicate either "Discharge Condition" or "Discharge Disposition". Our deterministic splitting algorithm maps common header variants to canonical section names and exploits the consistent relative ordering of sections to infer headers that remain ambiguous after redaction [8]. A residual 5,461 ambiguous cases could not be resolved automatically and were handled manually.

Information modeling

Tabular target schemas were developed iteratively in close consultation with clinicians. The difficulty is the structural and content diversity within individual sections: a Pertinent Results section ranges from a single-line negative culture report to multiple pages of microbiology panels and longitudinal imaging findings. Schemas therefore had to capture essential clinical depth without introducing hyper-specific fields that remain empty across most admissions, and had to stay concise enough for human verification. We analyzed source texts for recurring patterns to refine field definitions.

Three sections (history of present illness, brief hospital course, discharge instructions) vary too much between cases to admit a reliable field decomposition and are preserved verbatim.

Adaptive extraction

We use deterministic rule-based extraction (regular expressions and string matching) for simple cases and LLMs for complex narratives that require contextual understanding:

  • Rule-based: gender, services, discharge disposition
  • Hybrid: chief complaint, allergies, major procedures, discharge diagnosis, and discharge condition
  • LLM-only: past medical history, medications on admission, physical examination, pertinent results, discharge medications

LLM extraction used Gemini 2.5 Flash [5] and Gemini 3.1 Flash-Lite [6]. Complete prompts and generation settings are available in the code repository [8]. We provide the few-shot example files used for LLM extraction and the curated allergy standardization map in the extraction_resources/ directory. These resources contain sensitive clinical text and are therefore excluded from the code repository.

Privacy and de-identification

We used Gemini through the Google Cloud Vertex AI API with zero data retention. This follows PhysioNet’s requirements for using MIMIC data with LLMs and online services [7].

The discharge summaries were broadly de-identified by the original developers [2], with redactions applied conservatively to avoid retaining personal identifying information (PII), which also removed some clinically relevant information. We deliberately did not attempt to recover PII and only normalized MIMIC's redaction artifacts to [REDACTED].

Quality validation

Three medical doctors reviewed extractions side by side against the source text. Rule-based extractions were excluded from manual review because they are deterministic and auditable from the published code. Annotation was performed at the item level, which distinguishes a single incorrect field from an entirely incorrect sample.

For each section we sampled at random 171 admissions, split into 121 single-annotated and 50 dual-annotated cases. Overlap cases were independently assigned to two of the three medical doctors, which allows inter-annotator agreement to be measured while maximizing coverage. We report Gwet's AC1 [4] as a chance-corrected agreement metric. The annotators completed 2,210 case-level tasks comprising 13,744 item-level decisions.

Item-level accuracy ranges from 92.2% to 99.8% across sections, as shown in Table 1.

Table 1: Extraction accuracy (item-level) and inter-annotator agreement (Gwet's AC1) per section.
Section Accuracy (%) Gwet's AC1
Allergies 99.8 99.8
Chief Complaint 97.6 98.0
Discharge Condition 97.4 98.0
Discharge Diagnosis 97.1 95.8
Major Procedures 98.3 98.0
Past Medical History 95.2 90.1
Physical Examination 95.2 90.4
Pertinent Results 92.2 88.4
Discharge Medications 97.2 97.1
Medications on Admission 99.3 97.2

Data Description

The dataset covers 300,720 hospital admissions from 137,402 distinct patients, one discharge summary per admission. It comprises 16 CSV files holding 19.3 million rows in total.

Sixteen discharge summary sections are provided as verbatim free text. 13 of those are additionally extracted into 14 structured tables, where the physical examination section splits into a vitals table and a body systems table. The three text-only sections are history of present illness, brief hospital course, and discharge instructions.

Structure and Keys

Every table links to the hub table admissions through hadm_id. The admissions table contains subject_id, hadm_id, and note_id from the discharge summaries in MIMIC-IV-Note.

Six tables hold at most one row per admission and are keyed on hadm_id alone. These are admissions, discharge_sections, gender, services, discharge_disposition, and discharge_condition. Eight tables hold multiple rows per admission and are keyed on (hadm_id, seq_num), where seq_num preserves the order in which entries appear in the source section. The two physical examination tables are keyed on (hadm_id, timepoint) and (hadm_id, timepoint, system).

Files

Hubs

  • admissions.csv: the link table containing subject_id, hadm_id, and note_id
  • discharge_sections.csv: the verbatim text of all 16 sections after structural parsing and before any structured extraction

Section Tables

  • gender.csv: recorded sex (M or F)
  • services.csv: the service at discharge (enum over 20 services)
  • discharge_disposition.csv: the discharge destination (enum over 4 categories).
  • discharge_condition.csv: discharge condition, covering mental status, functional activity, and further clinical status fields
  • allergies.csv: lists allergies
  • chief_complaint.csv: lists chief complaints and context
  • discharge_diagnosis.csv: lists clinician-written primary and secondary discharge diagnoses
  • major_procedures.csv: lists major procedures
  • past_medical_history.csv: structured past medical history (prior conditions, family, substance, and functional history)
  • medications_on_admission.csv: admission medications including name, presentation form, dose, route, frequency, strength, duration, and PRN and hold flags
  • discharge_medications.csv: discharge medications structured like admission medications
  • physical_exam_vitals.csv: lists vital signs from physical examinations including timepoint (admission, discharge, interim, or unknown)
  • physical_exam_systems.csv: lists examined systems as an enum over 11 body systems, where status records normal, abnormal, or not assessed and abnormal_findings lists the documented findings. Includes the timepoint as well.
  • pertinent_results.csv: identified result blocks, each carrying a content type enum, the line range locating the block in the source section, per-type enum arrays, and a topic label for clinical notes

Data Dictionary

This dictionary describes all fields in the 16 CSV files of MIMIC-Note-Extract. Types correspond to the PostgreSQL loading schema: integer, real, text, varchar, boolean, enum, arrays, and jsonb.

admissions.csv

Fields in admissions.csv.
Field Type Description
subject_id integer MIMIC patient identifier.
hadm_id integer MIMIC hospital admission identifier; uniquely identifies a row in this table.
note_id varchar(25) Identifier of the source discharge summary in the MIMIC-IV-Note discharge table.

discharge_sections.csv

Fields in discharge_sections.csv.
Field Type Description
subject_id integer MIMIC patient identifier.
hadm_id integer Hospital admission identifier linking to admissions.
allergies text Raw text of the allergies section.
brief_hospital_course text Raw narrative describing the hospital course.
chief_complaint text Raw text describing the chief complaint and presentation context.
discharge_condition text Raw text describing the patient's condition at discharge.
discharge_diagnosis text Raw text listing discharge diagnoses.
discharge_disposition text Raw text recording the discharge destination or disposition.
discharge_instructions text Raw discharge instructions.
discharge_medications text Raw text listing medications at discharge.
gender text Raw text of the section recording sex.
history_of_present_illness text Raw narrative describing the history of present illness.
major_procedure text Raw text of the major surgical or invasive procedure section.
medications_on_admission text Raw text listing medications on admission.
past_medical_history text Raw text of the past medical history section, which may also contain family, substance, and functional history.
pertinent_results text Raw text of the pertinent results section; line numbers in pertinent_results.csv refer to this field.
physical_examination text Raw text of the physical examination section, including vital signs and body system assessments.
service text Raw text identifying the clinical service at discharge.

gender.csv

Recorded sex (M or F).

Fields in gender.csv.
Field Type Description
subject_id integer MIMIC patient identifier.
hadm_id integer Hospital admission identifier linking to admissions.
gender enum Recorded sex: M (male) or F (female).

services.csv

The service at discharge (enum over 20 services).

Fields in services.csv.
Field Type Description
subject_id integer MIMIC patient identifier.
hadm_id integer Hospital admission identifier linking to admissions.
service enum Clinical service label: MEDICINE, SURGERY, ORTHOPAEDICS, NEUROLOGY, CARDIOTHORACIC, NEUROSURGERY, OBSTETRICS/GYNECOLOGY, PSYCHIATRY, UROLOGY, PLASTIC, PODIATRY, OTOLARYNGOLOGY, [REDACTED], EMERGENCY, ANESTHESIOLOGY, DENTAL, RADIATION, OPHTHALMOLOGY, RADIOLOGY, or OP.

discharge_disposition.csv

The patient's discharge destination.

Fields in discharge_disposition.csv.
Field Type Description
subject_id integer MIMIC patient identifier.
hadm_id integer Hospital admission identifier linking to admissions.
discharge_disposition enum Discharge disposition: HOME (home), HOME_WITH_SERVICE (home with services), EXTENDED_CARE (extended care), or EXPIRED (death).

discharge_condition.csv

The patient's discharge condition, covering mental status, functional activity, and further clinical status fields.

Fields in discharge_condition.csv.
Field Type Description
subject_id integer MIMIC patient identifier.
hadm_id integer Hospital admission identifier linking to admissions.
mental_status enum Mental status at discharge: CLEAR (clear and coherent), CONFUSED_SOMETIMES, or CONFUSED_ALWAYS.
level_of_consciousness enum Level of consciousness: ALERT (alert and interactive), LETHARGIC_AROUSABLE, or LETHARGIC_NOT_AROUSABLE.
activity_status enum Mobility: INDEPENDENT (ambulatory independently), WITH_ASSISTANCE (ambulatory with assistance or an aid), CHAIR_ASSISTANCE (assistance to a chair or wheelchair), or BEDBOUND.
general_condition enum Overall clinical condition: STABLE, IMPROVED, WORSENED, GOOD, POOR, or CRITICAL.
pain_status enum Pain status: NO_PAIN, MODERATE_PAIN, or SEVERE_PAIN.
cardiovascular_status enum Hemodynamic status: STABLE or UNSTABLE.
vital_signs_status enum Qualitative assessment of vital signs: STABLE or UNSTABLE.
fever_status enum Fever status: NO_FEVER or FEVER_PRESENT.
nutrition_status enum Nutrition or feeding status: ORAL_DIET, TUBE_FEEDING, PARENTERAL_NUTRITION, or NPO (nothing by mouth).
wound_presence enum Presence of wounds or surgical incisions: NO_WOUNDS or WOUNDS_PRESENT.
wound_infection_status enum Signs of wound infection: NOT_INFECTED or INFECTED.
respiratory_status enum Respiratory support status: ROOM_AIR, OXYGEN_LOW_FLOW (low-flow oxygen), OXYGEN_HIGH_FLOW (high-flow oxygen or noninvasive ventilation), or MECHANICAL_VENTILATION.
orientation_status enum Orientation: ORIENTED_X3 (person, place, and time), ORIENTED_X2 (person and place), ORIENTED_X1 (person only), or DISORIENTED.
neurological_status enum Focal neurological deficits: FOCAL (present) or NO_FOCAL (absent).
patient_died boolean Whether patient died; defaults to false.
patient_left_ama boolean Whether patient left against medical advice; defaults to false.

allergies.csv

List of a patient's allergies.

Fields in allergies.csv.
Field Type Description
subject_id integer MIMIC patient identifier.
hadm_id integer Hospital admission identifier linking to admissions.
seq_num integer Identifier within the list, starting at 1.
title text Allergy entry; may contain an allergen name or a statement such as no known allergies.
title_std text Standardized allergen label.

chief_complaint.csv

List of a patient's chief complaints and the admission context.

Fields in chief_complaint.csv.
Field Type Description
subject_id integer MIMIC patient identifier.
hadm_id integer Hospital admission identifier linking to admissions.
seq_num integer Identifier within the list, starting at 1.
category enum Entry type: FINDING (presenting symptom or finding) or CONTEXT (context of presentation).
text text Presenting finding or context.

discharge_diagnosis.csv

List of clinician-written primary and secondary discharge diagnoses.

Fields in discharge_diagnosis.csv.
Field Type Description
subject_id integer MIMIC patient identifier.
hadm_id integer Hospital admission identifier linking to admissions.
seq_num integer Identifier within the list, starting at 1.
title text Diagnosis description.
is_primary boolean true for primary diagnoses; false for secondary diagnoses.

major_procedures.csv

List of a patient's major procedures.

Fields in major_procedures.csv.
Field Type Description
subject_id integer MIMIC patient identifier.
hadm_id integer Hospital admission identifier linking to admissions.
seq_num integer Identifier within the list, starting at 1.
title text Procedure description.
is_previous boolean true for previous procedures; false for current procedures.

past_medical_history.csv

List of a patient's past medical history (prior conditions, family, substance, and functional history).

Fields in past_medical_history.csv.
Field Type Description
subject_id integer MIMIC patient identifier of the admitted patient, including for family-history entries.
hadm_id integer Hospital admission identifier linking to admissions.
seq_num integer Identifier within the list, starting at 1.
subject enum Person to whom the history item refers: PATIENT or FAMILY.
family_relation text Family relationship for a family-history item.
concept_name text Name of the condition, intervention, substance-related item, or functional status.
category enum History category: CONDITION (disease, symptom, or behavioral event), INTERVENTION (procedure, surgery, or hospitalization), SUBSTANCE (substance history), or FUNCTIONAL_STATUS (mobility, disability, or dependency).
temporal_status enum Temporal classification: CURRENT (active, chronic, or present) or PAST (historical or resolved).

medications_on_admission.csv

List of a patient's admission medications including name, presentation form, dose, route, frequency, strength, duration, and PRN and hold flags.

Fields in medications_on_admission.csv.
Field Type Description
subject_id integer MIMIC patient identifier.
hadm_id integer Hospital admission identifier linking to admissions.
seq_num integer Identifier within the list, starting at 1.
name text Base medication name.
presentation_form enum Medication form or presentation unit, using the medication presentation vocabulary below.
dose_value real Numeric documented dose, interpreted together with dose_unit.
dose_unit enum Dose unit from the measurement or presentation vocabulary below, or [OTHER].
dose_redacted boolean true when the dose value is [REDACTED]; otherwise false.
route enum Administration route, using the medication route vocabulary below.
duration_value real Numeric treatment duration, interpreted together with duration_unit.
duration_unit enum Treatment duration unit from the period vocabulary below.
prn boolean Whether the medication is to be taken as needed (PRN); defaults to false.
prn_condition text Indication or condition for as-needed use.
hold boolean Whether the medication is held or subject to a hold instruction; defaults to false.
hold_condition text Condition under which to withhold the medication.
strength jsonb Medication potency or concentration, represented by the strength JSON structure below.
frequency jsonb Administration schedule, represented by one of the frequency JSON structures below.

discharge_medications.csv

List of a patient's discharge medications (structured like admission medications).

Fields in discharge_medications.csv.
Field Type Description
subject_id integer MIMIC patient identifier.
hadm_id integer Hospital admission identifier linking to admissions.
seq_num integer Identifier within the discharge medication list, starting at 1.
name text Base medication name.
presentation_form enum Medication form or presentation unit, using the medication presentation vocabulary below.
dose_value real Numeric documented dose, interpreted together with dose_unit.
dose_unit enum Dose unit from the measurement or presentation vocabulary below, or [OTHER].
dose_redacted boolean true when the dose value is [REDACTED]; otherwise false.
route enum Administration route, using the medication route vocabulary below.
duration_value real Numeric treatment duration, interpreted together with duration_unit.
duration_unit enum Treatment duration unit from the period vocabulary below.
prn boolean Whether the medication is to be taken as needed (PRN); defaults to false.
prn_condition text Indication or condition for as-needed use.
hold boolean Whether the medication is held or subject to a hold instruction; defaults to false.
hold_condition text Condition under which to withhold the medication.
strength jsonb Medication potency or concentration, represented by the strength JSON structure below.
frequency jsonb Administration schedule, represented by one of the frequency JSON structures below.

Medication controlled vocabularies

Vocabularies shared by both medication files and their JSON fields.
Vocabulary Allowed values
Presentation TABLET, CAPSULE, INHALER, PATCH, NEBULIZATION, APPLICATION, PUFF, DROP, DISK, SPRAY, OINTMENT, CREAM, GEL, LOZENGE, INJECTION, SOLUTION, [OTHER].
Measurement MG (milligram), MCG (microgram), G (gram), ML (milliliter), MEQ (milliequivalent), PERCENT (percent), UNIT (unit), or [OTHER].
Route ORAL, SUBCUTANEOUS, INHALATION, TOPICAL, OPHTHALMIC, SUBLINGUAL, NASAL, INJECTION, RECTAL, TRANSDERMAL, NEBULIZATION, MUCOSAL, or [OTHER].
Period MINUTE, HOUR, DAY, WEEK, MONTH, or [OTHER].
Specific time of day BEDTIME, BREAKFAST, LUNCH, DINNER, MORNING, AFTERNOON, EVENING, NIGHT, [OTHER], or [REDACTED].
Day of week MONDAY, TUESDAY, WEDNESDAY, THURSDAY, FRIDAY, SATURDAY, SUNDAY, or [REDACTED].

Medication strength JSON

Fields inside each medication strength JSON object.
Field JSON type Description
type string SINGLE for one numerator value, RANGE for two bounds, or COMBINATION for two or more component strengths.
numerator_values array of numbers or strings Strength values; a redacted value is represented by [REDACTED].
numerator_units array of strings Unit for each corresponding numerator value, from the medication measurement vocabulary.
denominator_value number, string, or null Denominator quantity, [REDACTED], or JSON null when no denominator is specified.
denominator_unit string or null Denominator unit from the medication measurement or presentation vocabulary, or JSON null.

Illustrative single-strength object:

{"type":"SINGLE","numerator_values":[10.0],"numerator_units":["MG"],"denominator_value":1.0,"denominator_unit":"TABLET"}

Medication frequency JSON

Fields inside each medication frequency JSON object, grouped by type.
Type value Field JSON type Description
All types type string One of times_per_period, every_period, specific_time, specific_day, or [OTHER].
times_per_period times integer or string Number of administrations per period, or [REDACTED].
times_per_period period_unit string Period unit from the medication period vocabulary.
every_period value integer or string Interval between administrations in the specified unit, or [REDACTED].
every_period unit string Interval unit from the medication period vocabulary.
specific_time time string Time label from the specific-time-of-day vocabulary.
specific_day days array of strings Scheduled days from the day-of-week vocabulary.

A frequency with type equal to [OTHER] contains no additional fields.

Illustrative frequency objects:

{"type":"times_per_period","times":2,"period_unit":"DAY"}
{"type":"every_period","value":8,"unit":"HOUR"}
{"type":"specific_time","time":"BEDTIME"}
{"type":"specific_day","days":["MONDAY","WEDNESDAY","FRIDAY"]}
{"type":"[OTHER]"}

physical_exam_vitals.csv

Lists a patient's vital signs from physical examinations.

Fields in physical_exam_vitals.csv.
Field Type Description
subject_id integer MIMIC patient identifier.
hadm_id integer Hospital admission identifier linking to admissions.
timepoint enum Examination timing: ADMISSION, DISCHARGE, INTERIM (during the stay), or UNKNOWN.
temperature_f real Temperature in degrees Fahrenheit.
temperature_redacted boolean true when the extracted temperature is [REDACTED]; otherwise false.
temperature_c real Temperature in degrees Celsius.
heart_rate integer Heart rate in beats per minute.
heart_rate_redacted boolean true when the extracted heart rate is [REDACTED]; otherwise false.
bp_systolic integer Systolic blood pressure in mmHg.
bp_systolic_redacted boolean true when the extracted systolic blood pressure is [REDACTED]; otherwise false.
bp_diastolic integer Diastolic blood pressure in mmHg.
bp_diastolic_redacted boolean true when the extracted diastolic blood pressure is [REDACTED]; otherwise false.
respiratory_rate integer Respiratory rate in breaths per minute.
o2_saturation integer Oxygen saturation in percent.
weight_value real Body weight, interpreted together with weight_unit.
weight_unit varchar(8) Weight unit: kg (kilograms) or lbs (pounds), when specified.
o2_support_method enum Oxygen or ventilatory support: ROOM_AIR, NASAL_CANNULA, HIGH_FLOW_NASAL_CANNULA, SIMPLE_MASK, NON_REBREATHER, CPAP_BIPAP (continuous or bilevel positive airway pressure), VENTILATOR, or AMBU_BAG (bag ventilation).
o2_support_lpm real Documented oxygen flow rate in liters per minute.
warnings text[] Array of physiological outlier warnings generated during postprocessing.

physical_exam_systems.csv

List of a patient's examined systems as an enum over 11 body systems, where status records normal, abnormal, or not assessed and abnormal_findings lists the documented findings. Includes the timepoint of the examination as well.

Fields in physical_exam_systems.csv.
Field Type Description
subject_id integer MIMIC patient identifier.
hadm_id integer Hospital admission identifier linking to admissions.
timepoint enum Examination timing: ADMISSION, DISCHARGE, INTERIM, or UNKNOWN
system enum Examined system: GENERAL, HEENT (head, eyes, ears, nose, and throat), CV (cardiovascular), RESP (respiratory), GI (gastrointestinal), EXTREM (extremities), NEURO (neurological), SKIN, PSYCH (psychiatric), GU (genitourinary), or ACCESS (clinical access sites and devices).
status enum Assessment category: NORMAL, ABNORMAL, or NOT_ASSESSED.
abnormal_findings text[] Array of extracted abnormal findings for the system.

pertinent_results.csv

List of a patient's pertinent results (lab tests, imaging studies, ECGs, etc.).

Fields in pertinent_results.csv.
Field Type Description
subject_id integer MIMIC patient identifier.
hadm_id integer Hospital admission identifier linking to admissions.
seq_num integer Position of the block in the section, starting at 1.
content_type enum Block category: LABORATORY_RESULTS, MICROBIOLOGY_RESULTS, IMAGING_STUDY, ELECTROPHYSIOLOGY_STUDY, PATHOLOGY_REPORT, or CLINICAL_NOTE.
start_line integer Inclusive starting line index in discharge_sections.pertinent_results, counting from 0.
end_line integer Inclusive ending line index in the same source section, counting from 0.
text_passage text Description of the pertinent result.
summary_topic text Brief topic label
lab_sample_types enum[] Sample categories for LABORATORY_RESULTS: PLASMA, SERUM, WHOLE_BLOOD, CEREBROSPINAL_FLUID, PERITONEAL_FLUID, URINE, STOOL, JOINT_FLUID, or UNKNOWN.
microbiology_types enum[] Categories for MICROBIOLOGY_RESULTS: BLOOD_CULTURE, URINE_CULTURE, SPUTUM_CULTURE, STOOL_CULTURE, WOUND_CULTURE, CSF_CULTURE, BODY_FLUID_CULTURE, CATHETER_TIP_CULTURE, VIRAL_PANEL_MOLECULAR, FUNGAL_STUDIES, PARASITOLOGY, ACID_FAST_MYCOBACTERIA, SENSITIVITY_REPORT, or UNKNOWN.
imaging_modalities enum[] Study or procedure categories for IMAGING_STUDY: CT, MRI, XR, FLUOROSCOPY, ULTRASOUND, ECHOCARDIOGRAM, CARDIAC_CATHETERIZATION, NUCLEAR_MEDICINE, PET, ESOPHAGOGASTRODUODENOSCOPY, COLONOSCOPY, ERCP, BRONCHOSCOPY, ANGIOGRAPHY_IR, MAMMOGRAPHY, or UNKNOWN.
body_parts enum[] Anatomic categories for IMAGING_STUDY: BRAIN, HEAD_NECK_EXCL_BRAIN, CHEST_THORAX, ABDOMEN_PELVIS, SPINE_ISOLATED, UPPER_EXTREMITY, LOWER_EXTREMITY, VASCULAR_SYSTEM, WHOLE_BODY, or UNKNOWN.
electrophysiology_types enum[] Categories for ELECTROPHYSIOLOGY_STUDY: EEG, ECG, AMBULATORY_ECG, EMG, NERVE_CONDUCTION, STRESS_TEST, SLEEP_STUDY, PACEMAKER_INTERROGATION, or UNKNOWN.
pathology_types enum[] Categories for PATHOLOGY_REPORT: SURGICAL_PATHOLOGY, CYTOPATHOLOGY, AUTOPSY, BONE_MARROW, or UNKNOWN.

Usage Notes

Raw CSVs

The extracted discharge summary sections are provided as gzipped CSV files.

Note: discharge_sections.csv contains verbatim clinical text with embedded newlines within quoted fields. CSV readers must therefore support quoted multiline fields.

PostgreSQL

We recommend loading the files into PostgreSQL using the scripts in our code repository [8], which supply typed columns, enums, primary keys, and foreign keys to admissions. Extract the CSV files, then run create.sql to define the schema and load.sql to load the data.

Croissant

We also provide Croissant 1.1 metadata [9], which describes the files, column types, keys, foreign keys to admissions, and controlled vocabularies. The metadata can be used directly with the published gzip files.

Intended use cases

The database supports a broad range of applications by making clinical information from discharge summaries available in structured tables that can be linked to the other MIMIC-IV tables. Use case examples include:

  • Cohort identification: Selecting admissions based on documented clinical histories, findings, or treatments.
  • Medication reconciliation: Comparing pre-admission and discharge medications with inpatient prescription records.
  • Allergy-related prescribing research: Studying associations between reported allergy labels, antibiotic prescribing, and clinical outcomes.
  • Risk adjustment: Incorporating documented medical history and functional status into outcome analyses.
  • Clinical decision-making benchmarks: Constructing patient cases for developing and evaluating diagnostic models and agents.

Potential limitations

The database has limitations arising from its extraction methods and source data:

  • Extraction errors: Section splitting and rule-based or LLM-based extraction may have introduced errors. Clinician review of sampled extractions supports the extractions' quality but cannot establish correctness for every record.
  • Incomplete representation: Predefined schemas do not capture all information in the original summaries, and some information was intentionally omitted during schema design. The schemas were iteratively refined with clinicians to retain clinically relevant content. Although unlikely, important details may still have been overlooked. Missing entries should not be interpreted as evidence that a condition or finding was absent.
  • Cohort selection: Excluding incomplete or unusually formatted summaries may have introduced selection bias.
  • Source-data limitations: The database inherits MIMIC-IV-Note's single-center coverage, incomplete or inaccurate clinical documentation, and information loss from de-identification.

Release Notes

Version 1.0.0

Initial release

Ethics

The dataset is a derivative dataset of MIMIC-IV and thus no new patient data was collected. The ethics approval of the dataset follows from that of the parent MIMIC-IV-Note dataset.

Conflicts of Interest

No conflicts of interest.

References

  1. Johnson, A., Bulgarelli, L., Pollard, T., Horng, S., Celi, L. A., & Mark, R. (2023). MIMIC-IV (version 2.2). PhysioNet. RRID:SCR_007345. https://doi.org/10.13026/6mm1-ek67
  2. Johnson, A., Pollard, T., Horng, S., Celi, L. A., & Mark, R. (2023). MIMIC-IV-Note: Deidentified free-text clinical notes (version 2.2). PhysioNet. RRID:SCR_007345. https://doi.org/10.13026/1n74-ne17
  3. Zhang, X., Kong, W., Shi, R., Sun, L., Xu, M., & Gong, L. (2025). Data-driven trends in critical care informatics: a bibliometric analysis of global collaborations using the MIMIC database (2004-2024). Computers in Biology and Medicine, 195, 110670. https://doi.org/10.1016/j.compbiomed.2025.110670
  4. Gwet, K. L. (2008). Computing inter-rater reliability and its variance in the presence of high agreement. British Journal of Mathematical and Statistical Psychology, 61(1), 29-48. https://doi.org/10.1348/000711006X126600
  5. Google DeepMind. Gemini 2.5 Flash, 2025. URL https://storage.googleapis.com/deepmind-media/Model-Cards/Gemini-2-5-Flash-Model-Card.pdf Accessed: 16.06.26
  6. Google DeepMind. Gemini 3.1 Flash Lite, 2026. URL https://deepmind.google/models/model-cards/gemini-3-1-flash-lite/ Accessed: 16.06.26
  7. Use of MIMIC data with large language models and online services. https://physionet.org/news/post/llm-responsible-use/ Accessed: 27.07.26
  8. mimic-note-extract. https://github.com/kschwethelm/mimic-note-extract/ Accessed: 20.10.26
  9. MLCommons. Croissant format specification. https://docs.mlcommons.org/croissant/docs/croissant-spec.html Accessed: 28.07.26
Contents
Parent Projects
MIMIC-IV-Ext MIMIC Note Extract: A Large-Scale Structured Dataset Extracted from MIMIC-IV-Note Discharge Summaries was derived from: Please cite them when using this project.
Share
Access
Access Policy
Only credentialed users who sign the DUA can access the files.
License
PhysioNet Credentialed Health Data License 1.5.0
Data Use Agreement
PhysioNet Credentialed Health Data Use Agreement 1.5.0
Required training
CITI Data or Specimens Only Research
Discovery
DOI (v1.0)
10.13026/cq50-rd11
DOI (latest)
10.13026/rpd5-e431
Topics
clinical nlp mimic large language models electronic health records clinical information extraction structured clinical datasets
Project Views
Current Version
4
All Versions
4
View Details
Corresponding Author
You must be logged in to view the contact information.
Versions

Files