Table 7.

Examples of extracted EI compared with Orphanet data

Diseases	EI labels	BioBERT EI extractor	Orphanet epidemiologic data
Rett syndrome (GARD:0005696), a neurodevelopmental disorder that is characterized by developmental delay and regression, abnormal respiration, absent speech, and inconsolable screaming, crying, panic-like attacks, and gnashing of teeth [84]	Key Phrase from Abstract	“Five patients with definite RS were identified in a population of 203,801 children (98,932 girls) 0–18 years of age yielding a prevalence rate of RS of 1 in 40,760 in North Dakota children.” [85]
	LOC	North Dakota	United States
	EPI	Prevalence rate(s)	Prevalence at birth
	STAT	1 in 40,760, which can be normalized to 2.45 per 100,000	1–9/100,000
Eosinophilic esophagitis (GARD:0009142), characterized by nausea, vomiting, and pain from inflammation in the esophagus caused by eosinophil invasion [86]	Key Phrase from Abstract	“…series of EE have also been reported in Japan. … The prevalence of EE was calculated to be 17.1/100,000.” [87]
	LOC	Japan	Japan
	EPI	Prevalence	Point prevalence
	STAT	17.1/100,000	17.1 per 100,000
Smith-Magenis syndrome (GARD:0008197), is “associated with psychomotor delay, a particular behavioural pattern and congenital anomalies.”[88]	Key Phrase from Abstract	“Smith-Magenis syndrome (SMS) is rare (prevalence 1 in 25 000)” [88]
	LOC	Worldwide	Worldwide
	EPI	Prevalence	Point prevalence
	STAT	1 in 25 000, which is normalized to 4 per 100,000	4 per 100,000
Fibrodysplasia ossificans progressiva (GARD:0006445), an autosomal dominant disorder characterized by a single mutation that leads to the painful ossification of skeletal muscle, tendons, and ligaments after trauma [89]	Key Phrase from Abstract	“Previous studies found that the FOP prevalence was about one in every two million lives. The aim of this study is to estimate the FOP prevalence in France” “89 FOP patients were identified, which results in a prevalence of 1.36 per million inhabitants (CI95% = [1.10; 1.68]).”[90]
	LOC	France	France
	EPI	Prevalence	Point prevalence
	STAT	1.36 per million inhabitants, which can be normalized to 0.136 per 100,000, and about one in every two million lives	0.136 per 100,000
Granulomatosis with polyangiitis (GARD:0007880), also called Wegener granulomatosis, is an autoimmune disease characterized by sinus and joint pain, respiratory infections, and skin lesions caused by inflammation of blood vessels [91]	Key Phrase from Abstract	“annual incidence/ million population increased from 5.2 (95% confidence interval [95% CI] 2.7–9.0) during 1984–1988 to 12.0 (95% CI 8.0–17.3) during 1994–1998. The point prevalence/million increased from 30.4 (95% CI 16.6–51.0) to 95.1 (95% CI 69.1–129.0).”[92]
	LOC	Northern Norway	Norway
	EPI	Incidence, incidence rate, annual incidence, prevalence, prevalence rates, point prevalence	Point prevalence
	STAT	Million population, million, 30. 4, to 95. 1	9.51 per 100,000