Table 1 Some of the previous i2b2 challenge tasks involving identifying risk factors for heart disease in clinical notes.

i2b2 de-identification and smoking challenge (2006)

Automatic identification of patient smoking status and de-identification of personal health information

De-identification: 0.98;

Smoking identification: 0.90

^{49, 54}

i2b2 obesity challenge (2008)

Identification of obesity and its co-morbidities

0.9773

⁵⁰

i2b2 medication challenge (2009)

Identification of medications, their dosages, administration methods, frequencies, durations, and administration reasons from discharge summaries

Durations identification:0.525;

Reason identification:0.459

⁴⁸

i2b2 relations challenge (2010)

Concept extraction, and classification of assertion and relation

Concept extraction: 0.852;

Classification of assertion and relation: 0.936

⁵¹

i2b2 coreference challenge (2011)

Coreference resolution

0.827

⁵⁵

i2b2 temporal relations challenge (2012)

Extraction of temporal relations from clinical records involving identification of temporal expressions, temporal relations, and significant clinical events

Event: 0.92;

Temporal expression: 0.90;

Temporal relation: 0.69

⁵²

i2b2 de-identification and heart disease risk factors challenge (2014)

Automatic de-identification and identification of CAD risk factors in the narratives of diabetes patients’ longitudinal clinical records

De-identification: 0.9586;

Risk factor: 0.9276

^{56, 57}

CLEF eHealth shared task 1 (2013)

Named entity recognition in clinical notes

0.75

⁵⁸

CLEF eHealth shared task 1b (2014)

Normalization of abbreviations or acronyms

Task 2a: 0.868 (accuracy);

Task 2b: 0.576 (F-measure)

⁵⁹

CLEF eHealth shared Evaluation (2020)

Clinical named entity recognition from French clinical notes

Recognition of plain entity: 0.756;

Recognition of normalized entity: 0.711;

Entity normalization: 0.872

⁶⁰

CLEF eHealth shared Evaluation (2021)

Clinical named entity recognition from French medical text

Recognition of plain entity: 0.702;

Recognition of normalized entity: 0.529;

j Entity normalization: 0.524

⁶¹

SemEval task 9 (2013)

Extraction of drug-drug interactions from clincial texts

Drugs recognition: 0.715;

Drug-drug interactions extraction: 0.651

⁶²

SemEval task 7 (2014)

Identification and normalization of diseases and disorders in clinical notes

Identification: 0.813;

Normalization: 0.741 (accuracy)

⁶³

SemEval task 14 (2015)

Named entity recognition and filling template slot for clinical notes

Named entity recognition: 0.757;

Template slot filling accuracy:0.886;

Recognition of disorder and template slot filling accuracy: 0.808

⁶⁴

SemEval task (2016)

Extraction of temporal information from clinical notes involving identification of time expression, event expression and temporal relation

Identification of time expression: 0.795;

Identification of event expression: 0.903;

Identification of temporal relation: 0.573

⁴⁶