DDXPlus: A New Dataset For Automatic Medical Diagnosis

Appearing in NeurIPS 2022 dataset and benchmark track

We are releasing under the CC-BY licence a new large-scale dataset for Automatic Symptom Detection (ASD) and Automatic Diagnosis (AD) systems in the medical domain.

The dataset contains patients synthesized using a proprietary medical knowledge base and a commercial rule-based ASD system. Patients in the dataset are characterized by their socio-demographic data, a pathology they are suffering from, a set of symptoms and antecedents related to this pathology, and a differential diagnosis. The symptoms and antecedents can be binary, categorical and multi-choice, with the potential of leading to more efficient and natural interactions between ASD/AD systems and patients.

To the best of our knowledge, this is the first large-scale dataset that includes the differential diagnosis, and non-binary symptoms and antecedents.

• DDXPlus: A New Dataset For Automatic Medical Diagnosis
  • Availability
  • Dataset documentation
    • Evidence description
      • Example
    • Pathology description
      • Example
    • Patient description
      • Example
  • Dataset statistics
    • Pathology statistics
    • Socio-demographic statistics
    • Distribution of the evidence types
    • Number of evidences of the synthesized patients
    • Differential diagnosis statistics
  • Experiments

Availability

• Our paper is available on arXiv.
• The dataset in French is hosted on figshare.
  • This is the original version of DDXPlus that all results in our paper were obtained on.
• Starting from 9 May 2023, the dataset is also available in English for easier use. This version is hosted on figshare.
  • The English version of DDXPlus contains the same data in the same format as the French version.
  • Wherever possible, English names or non-semantic codes are used instead of French names.
  • Using the English version should lead to the same performance as using the French version.

Looking for a text version?

We're hoping to release a text version of DDXPlus in the coming weeks/months! Meanwhile, you can check out this script to convert it into conversational format.

Dataset documentation

In what follows, we use the term evidence as a general term to refer to a symptom or an antecedent. The dataset contains the following files:

• release_evidences.json: a JSON file describing all possible evidences considered in the dataset.
• release_conditions.json: a JSON file describing all pathologies considered in the dataset.
• release_train_patients.zip: a CSV file containing the patients of the training set.
• release_validate_patients.zip: a CSV file containing the patients of the validation set.
• release_test_patients.zip: a CSV file containing the patients of the test set.

Evidence description

Each evidence in the release_evidences.json file is described using the following entries:

• name: name of the evidence.
  • In the English version, this is replaced with a unique, non-semantic code starting with E.
• code_question: a code allowing to identify which evidences are related. Evidences having the same code_question form a group of related symptoms. The value of the code_question refers to the evidence that need to be simulated/activated for the other members of the group to be eventually simulated.
• question_fr: the query, in French, associated to the evidence.
• question_en: the query, in English, associated to the evidence.
• is_antecedent: a flag indicating whether the evidence is an antecedent or a symptom.
• data_type: the type of the evidence. We use "B" for binary, "C" for categorical, and "M" for multi-choice.
• default_value: the default value of the evidence. If this value is used to characterize the evidence, then it is as if the evidence was not synthesized.
• possible-values: the possible values for the evidence. Only valid for categorical and multi-choice evidences.
  • In the English version, every value is replaced with a unique, non-semantic code starting with V.
• value_meaning: The meaning, in French and English, of each code that is part of the possible-values field. Only valid for categorical and multi-choice evidences.

Example

English

{
    "name": "E_130",
    "code_question": "E_129",
    "question_fr": "De quelle couleur sont les lésions?",
    "question_en": "What color is the rash?",
    "is_antecedent": false,
    "default_value": "V_11",
    "value_meaning": {
        "V_11": {"fr": "NA", "en": "NA"},
        "V_86": {"fr": "foncée", "en": "dark"},
        "V_107": {"fr": "jaune", "en": "yellow"},
        "V_138": {"fr": "pâle", "en": "pale"},
        "V_156": {"fr": "rose", "en": "pink"},
        "V_157": {"fr": "rouge", "en": "red"}
    },
    "possible-values": [
        "V_11",
        "V_86",
        "V_107",
        "V_138",
        "V_156",
        "V_157"
    ],
    "data_type": "C"
}

French

{
    "name": "lesions_peau_couleur",
    "code_question": "lesions_peau",
    "question_fr": "De quelle couleur sont les lésions?",
    "question_en": "What color is the rash?",
    "is_antecedent": false,
    "default_value": "NA",
    "value_meaning": {
        "NA": {"fr": "NA", "en": "NA"},
        "foncee": {"fr": "foncée", "en": "dark"},
        "jaune": {"fr": "jaune", "en": "yellow"},
        "pale": {"fr": "pâle", "en": "pale"},
        "rose": {"fr": "rose", "en": "pink"},
        "rouge": {"fr": "rouge","en": "red"}
    },
    "possible-values": [
        "NA",
        "foncee",
        "jaune",
        "pale",
        "rose",
        "rouge"
    ],
    "data_type": "C"
}

Pathology description

The file release_conditions.json contains information about the pathologies patients in the datasets may suffer from. Each pathology has the following attributes:

• condition_name: name of the pathology.
  • In the English version, the English name is used instead of the French name.
• cond-name-fr: name of the pathology in French.
• cond-name-eng: name of the pathology in English.
• icd10-id: ICD-10 code of the pathology.
• severity: the severity associated with the pathology. The lower the more severe.
• symptoms: data structure describing the set of symptoms characterizing the pathology. Each symptom is represented by its corresponding name entry in the release_evidences.json file.
• antecedents: data structure describing the set of antecedents characterizing the pathology. Each antecedent is represented by its corresponding name entry in the release_evidences.json file.

Example

English

{
    "condition_name": "Myasthenia gravis",
    "cond-name-fr": "Myasthénie grave",
    "cond-name-eng": "Myasthenia gravis",
    "icd10-id": "G70.0",
    "symptoms": {
        "E_65": {},
        "E_63": {},
        "E_52": {},
        "E_172": {},
        "E_84": {},
        "E_66": {},
        "E_90": {},
        "E_38": {},
        "E_176": {}
     },
    "antecedents": {
        "E_28": {},
        "E_204": {}
    },
    "severity": 3
}

French

{
    "condition_name": "Myasthénie grave",
    "cond-name-fr": "Myasthénie grave",
    "cond-name-eng": "Myasthenia gravis",
    "icd10-id": "G70.0",
    "symptoms": {
        "dysphagie": {},
        "dysarthrie": {},
        "diplopie": {},
        "ptose": {},
        "faiblesse_msmi": {},
        "dyspn": {},
        "fatigabilité_msk": {},
        "claud_mâchoire": {},
        "rds_paralys_gen": {}
    },
    "antecedents": {
        "atcdfam_mg": {},
        "trav1": {}
    },
    "severity": 3
}

Patient description

Each patient in each of the 3 sets has the following attributes:

• AGE: the age of the synthesized patient.
• SEX: the sex of the synthesized patient.
• PATHOLOGY: name of the ground truth pathology (cf condition_name property in the release_conditions.json file) that the synthesized patient is suffering from.
• EVIDENCES: list of evidences experienced by the patient. An evidence can either be binary, categorical or multi-choice. A categorical or multi-choice evidence is represented in the format [evidence-name]_@_[evidence-value] where [evidence-name] is the name of the evidence (name entry in the release_evidences.json file) and [evidence-value] is a value from the possible-values entry. Note that for a multi-choice evidence, it is possible to have several [evidence-name]_@_[evidence-value] items in the evidence list, with each item being associated with a different evidence value. A binary evidence is represented as [evidence-name].
• INITIAL_EVIDENCE: the evidence provided by the patient to kick-start an interaction with an ASD/AD system. This is useful during model evaluation for a fair comparison of ASD/AD systems as they will all begin an interaction with a given patient from the same starting point. The initial evidence is randomly selected from the evidence list mentioned above (i.e., EVIDENCES) and it is part of this list.
• DIFFERENTIAL_DIAGNOSIS: The ground truth differential diagnosis for the patient. It is represented as a list of pairs of the form [[patho_1, proba_1], [patho_2, proba_2], ...] where patho_i is the pathology name (condition_name entry in the release_conditions.json file) and proba_i is its related probability.

Example

English

{
    "AGE": 18,
    "DIFFERENTIAL_DIAGNOSIS": [["Bronchitis", 0.19171203430383882], ["Pneumonia", 0.17579340398940366], ["URTI", 0.1607809719801254], ["Bronchiectasis", 0.12429044460990353], ["Tuberculosis", 0.11367177304035844], ["Influenza", 0.11057936110639896], ["HIV (initial infection)", 0.07333003867293564], ["Chagas", 0.04984197229703562]],
    "SEX": "M",
    "PATHOLOGY": "URTI",
    "EVIDENCES": ["E_48", "E_50", "E_53", "E_54_@_V_161", "E_54_@_V_183", "E_55_@_V_89", "E_55_@_V_108", "E_55_@_V_167", "E_56_@_4", "E_57_@_V_123", "E_58_@_3", "E_59_@_3", "E_77", "E_79", "E_91", "E_97", "E_201", "E_204_@_V_10", "E_222"],
    "INITIAL_EVIDENCE": "E_91"
}

French

{
    "AGE": 18, 
    "DIFFERENTIAL_DIAGNOSIS": [["Bronchite", 0.19171203430383882], ["Pneumonie", 0.17579340398940366],["IVRS ou virémie", 0.1607809719801254], ["Bronchiectasies", 0.12429044460990353], ["Tuberculose", 0.11367177304035844], ["Possible influenza ou syndrome virémique typique", 0.11057936110639896], ["VIH (Primo-infection)", 0.07333003867293564], ["Chagas", 0.04984197229703562]], 
    "SEX": "M", 
    "PATHOLOGY": "IVRS ou virémie", 
    "EVIDENCES": ["crowd", "diaph", "douleurxx", "douleurxx_carac_@_sensible", "douleurxx_carac_@_une_lourdeur_ou_serrement", "douleurxx_endroitducorps_@_front", "douleurxx_endroitducorps_@_joue_D_", "douleurxx_endroitducorps_@_tempe_G_", "douleurxx_intens_@_4", "douleurxx_irrad_@_nulle_part", "douleurxx_precis_@_3", "douleurxx_soudain_@_3", "expecto", "f17.210", "fievre", "gorge_dlr", "toux", "trav1_@_N", "z77.22"], 
    "INITIAL_EVIDENCE": "fievre"
}

Dataset statistics

Pathology statistics

Socio-demographic statistics

Distribution of the evidence types

	Binary	Categorical	Multi-choice	Total
Evidences	208	10	5	223
Symptoms	96	9	5	110
Antecedents	112	1	0	113

Number of evidences of the synthesized patients

	Avg	Std dev	Min	1st quartile	Median	3rd quartile	Max
Evidences	13.56	5.06	1	10	13	17	36
Symptoms	10.07	4.69	1	8	10	12	25
Antecedents	3.49	2.23	0	2	3	5	12

Differential diagnosis statistics

Experiments

Code for reproducing results in the paper can be found in code.

In our paper, we reported results of two methods, a RL-based method AARLC and a supervised method BASD which is adapted from ASD. For instructions on how to run them, see here for AARLC and here for BASD.